SAforArcGIS Web

Geographic Imaging by Leica Geosystems GIS & Mapping

Using Stereo Analyst for ArcGIS

Kris Curry

Using Stereo Analyst for ArcGIS

Copyright © 2003 Leica Geosystems GIS & Mapping, LLCAll rights reserved.Printed in the United States of America.

The information contained in this document is the exclusive property of Leica Geosystems GIS & Mapping, LLC. This work is protected under United States copyright law and other international copyright treaties and conventions. No part of this work may be reproduced or transmitted in any form or by any means, electronic or mechanical, including photocopying and recording, or by any information storage or retrieval system, as expressly permitted in writing by Leica Geosystems GIS & Mapping, LLC. All requests should be sent to Attention: Manager of Technical Documentation, Leica Geosystems GIS & Mapping, LLC, 2801 Buford Highway NE, Suite 400, Atlanta, GA, 30329-2137, USA.

The information contained in this document is subject to change without notice.

CONTRIBUTORS

Contributors to this book and On-line Help for Stereo Analyst for ArcGIS include: Sam Megenta, Frank Obusek, Jay Pongonis, Russ Pouncey, Mladen Stojic′, Ryan Strynatka, and Lori Zastrow of Leica Geosystems GIS & Mapping, LLC.

U.S. GOVERNMENT RESTRICTED/LIMITED RIGHTS

Any software, documentation, and/or data delivered hereunder is subject to the terms of the License Agreement. In no event shall the U.S. Government acquire greater than RESTRICTED/LIMITED RIGHTS. At minimum, use, duplication, or disclosure by the U.S. Government is subject to restrictions set forth in FAR §52.227-14 Alternates I, II, and III (JUN 1987); FAR §52.227-19 (JUN 1987), and/or FAR §12.211/12.212 (Commercial Technical Data/Computer Software); and DFARS §252.227-7015 (NOV 1995) (Technical Data) and/or DFARS §227.7202 (Computer Software), as applicable. Contractor/Manufacturer is Leica Geosystems GIS & Mapping, LLC, 2801 Buford Highway NE, Suite 400, Atlanta, GA, 30329-2137, USA.

ERDAS, ERDAS IMAGINE, IMAGINE OrthoBASE, and Stereo Analyst for ArcGIS are registered trademarks. IMAGINE OrthoBASE Pro and IMAGINE VirtualGIS are trademarks.

SOCET SET is a registered trademark of BAE Systems Mission Solutions.

ERDAS® is a wholly owned subsidiary of Leica Geosystems GIS & Mapping, LLC.

Other companies and products mentioned herein are trademarks or registered trademarks of their respective trademark owners.

ContentsContents

Contents iii

D 31

76ages 78

Foreword vii

Getting started

1 Introducing Stereo Analyst for ArcGIS 3What can you do with Stereo Analyst for ArcGIS? 4Learning about Stereo Analyst for ArcGIS 11

2 Quick-start tutorial 13Exercise 1: Starting Stereo Analyst for ArcGIS 14Exercise 2: Adding oriented images 18Exercise 3: Converting features—3D to 2D and 2D to 3Exercise 4: Collecting features in 3D 42Exercise 5: Editing existing features 58What’s next? 66

Working in stereo

3 Working with oriented images 69Creating oriented images 70Using IMAGINE OrthoBASE to create oriented imagesUsing Image Analysis for ArcGIS to create oriented imImporting IMAGINE OrthoBASE block files 79Importing SOCET SET® files 82What’s next? 85

III

4 Working with 3D data 87Comparing 3D features and 3D models 88

Using Virtual 2D To 3D 89
Setting Virtual 2D To 3D options 90Using the 2D to 3D converter 93Using advanced conversion options 95Using the 3D to 2D converter 103What’s next? 104
5 Visualizing in stereo 105Introducing stereo visualization 106Using Stereo Window views 108Learning about the Stereo Analyst toolbar 111Learning about the Stereo View toolbar 112Learning about the Stereo Enhancement toolbar 113Using Stereo Analyst for ArcGIS with ArcMap 119Setting stereo display options 124What’s next? 129

6 Applying the 3D Floating Cursor 131Using the 3D Floating Cursor 132Adjusting the position of the 3D Floating Cursor 133Selecting 3D Floating Cursor options 136Using the Terrain Following Mode 138Applying other Terrain Following Mode options 141Checking accuracy of 3D information 146Using the 3D Floating Cursor 147Using keyboard shortcuts 149What’s next? 151

USING STEREO ANALYST FOR ARCGISIV

Working with features

7 Capturing GIS data 155Collecting features in different modes 156Using 3D Snap 162Using Squaring 167Using the Monotonic Mode 172Using digitizing devices 173What’s next? 176

Appendices

A Capturing data using imagery 179Collecting data for a GIS 180Preparing imagery for a GIS 182Using traditional approaches 186Applying geographic imaging 188Moving from imagery to a 3D GIS 190Identifying workflow 191Getting 3D GIS data from imagery 195

B Understanding stereo viewing 199Learning principles of stereo viewing 200Understanding stereo models and parallax 202Understanding scaling, translation, and rotation 205Understanding the epipolar line 206

C Applying photogrammetry 209

CONTENTS V

Learning principles of photogrammetry 210Acquiring images and data 214

Scanning aerial photography 216
Understanding interior orientation 223Understanding exterior orientation 226Using digital mapping solutions 229
Glossary 233

References 249

Index 251

USING STEREO ANALYST FOR ARCGISVI

VII

Fore

s capture a tains, and other ply recording ccur in the real shots of reality. record a specific ing cities, rivers,

need to be dy, easy-to-use in creating and y. Today’s features, es.

Where, What, done within the where is that is that? The new use imagery to

o you can then

trial usage and aphy changes, so

word

An image of the earth’s surface is a wealth of information. Imagepermanent record of buildings, roads, rivers, trees, schools, mounfeatures located on the earth’s surface. But images go beyond simfeatures. Images also record relationships and processes as they oworld. Images are snapshots of geography, but they are also snapImages chronicle our earth and everything associated with it; theyplace at a specific point in time. They are snapshots of our changand mountains. Images are snapshots of life on earth.

The data in a GIS needs to reflect reality, and snapshots of realityincorporated and accurately transformed into instantaneously reainformation. From snapshots to digital reality, images are pivotalmaintaining the information infrastructure used by today’s societgeographic information systems have been carefully created withattributed behavior, analyzed relationships, and modeled process

There are five essential questions that any GIS needs to answer: When, Why, and How. Uncovering Why, When, and How are allGIS; images allow you to extract the Where and What. Preciselybuilding? What is that parcel of land used for? What type of tree extensions developed by Leica Geosystems GIS & Mapping, LLCallow you to accurately address the questions Where and What, sderive answers for the other three.

But our earth is changing! Urban growth, suburban sprawl, indusnatural phenomena continually alter our geography. As our geogr

LYST FOR ARCGIS

does the information we need to understand it. Because an

USING STEREO ANAVIII

image is a permanent record of features, behavior, relationships, and processes captured at a specific moment in time, using a series of images of the same area taken over time allows you to more accurately model and analyze the relationships and processes that are important to our earth.

The new extensions by Leica Geosystems are technological breakthroughs which allow you to transform a snapshot of geography into information that digitally represents reality in the context of a GIS. Image Analysis™ for ArcGIS and Stereo Analyst® for ArcGIS are tools built on top of a GIS to maintain that GIS with up-to-date information. The extensions provided by Leica Geosystems reliably transform imagery directly into your GIS for analyzing, mapping, visualizing, and understanding our world.

On behalf of the Image Analysis for ArcGIS and Stereo Analyst for ArcGIS product teams, I wish you all the best in working with these new products and hope you are successful in your GIS and mapping endeavors.

Sincerely,

Mladen Stojic′Product ManagerLeica Geosystems GIS & Mapping, LLC

Get

ion 1

ting started

Sect

LYST FOR ARCGIS
USING STEREO ANA2

3

1Introducing

ction extension ing and feature

g feature editing

re datasets tasets accurately ve some feature

ng tools.

e created with information. By n more ion system (GIS) l (2D) GIS data. can collect mass errain model

IN THIS C

• What caAnalyst

• LearningArcGIS

1Intrfor A

Stereo Analyst for ArcGIS

Welcome to Stereo Analyst® for ArcGIS, the stereo feature collefor ArcGIS™. Stereo Analyst for ArcGIS adds unique image viewcollection capabilities to your ArcGIS desktop and uses the existinand collection capabilities available in ArcMap™.

With Stereo Analyst for ArcGIS, you can access image and featudirectly from a geodatabase. You can also collect new feature dausing oriented imagery as a reference backdrop. If you already hadatasets, you can edit them reliably in stereo using ArcMap editi

Using a three-dimensional (3D) digital view of the earth’s surfacoriented imagery, you can collect true, real-world 3D geographicanalyzing oriented imagery with Stereo Analyst for ArcGIS, eveinformation can be extracted from imagery. Geographic informatprofessionals are no longer limited to collecting two-dimensionaThis data can be used to build relationships into a GIS. Also, you points with X, Y, and Z coordinates for the creation of a digital t(DTM).

HAPTER

n you do with Stereo for ArcGIS?

about Stereo Analyst for

oducing Stereo AnalystrcGIS

LYST FOR ARCGIS

What can you do wi th Stereo Analyst for ArcGIS?ciated with it. As sor, a new 3D f GIS feature data loating Cursor is llected. When the of interest, a new an be accurately Floating Cursor” Floating Cursor

tereo Analyst for eographic feature , is used to (See “Using the ation about the

4

Stereo to updaa GIS. two appArcCatmanageAnalys

Crea

Stereo represeUsing Grepreseis displ

By addsuperimdigital featuregeograinform

Gaincolle

DuringAnalysthe 3D screen.point. (page 13

The 3Dinterestthis 3Dabove,

USING STEREO ANA

ing more accuracy in GIS data ction

the collection of accurate GIS feature data using Stereo t for ArcGIS, a unique 3D Floating Cursor “floats” within digital depiction of the earth’s surface as displayed on the This 3D Floating Cursor is also referred to as a ground See chapter 6, “Applying the 3D Floating Cursor” on 1 for more information about the 3D Floating Cursor.)

Floating Cursor floats until you place it on the feature of . You can move the 3D Floating Cursor anywhere within digital representation. The 3D Floating Cursor can float below, or rest on the earth’s surface or feature of interest.

This IMAGINE VirtualGIS™ scene shows a 3D world.

Automated techniques have been developed in SArcGIS to place the 3D Floating Cursor on the gof interest. This feature, Terrain Following Modefacilitate accurate feature collection and editing. Terrain Following Mode” on page 138 for informTerrain Following Cursor.)

Analyst for ArcGIS supplies you with the tools you’ll need te and create accurate and reliable feature datasets for use in You can use Stereo Analyst for ArcGIS in conjunction with lications you’re probably already familiar with,

alog™ and ArcMap. These applications allow you to easily all of your raster and feature datasets used in Stereo

t for ArcGIS.

t ing your world in 3D

Analyst for ArcGIS creates an accurate 3D digital ntation of the earth’s surface and geography using imagery. IS-ready images, the contents of an image are recreated and

nted in a 3D view. This 3D digital representation of the earth ayed on the screen.

ing feature datasets to ArcMap, the features are posed on top of the 3D digital representation. Using the 3D

picture of the earth’s surface as a reference source, the data is updated to “fit” the earth and its associated phy. Once updates to the feature data have been made, the ation can be saved or checked back into the geodatabase.

The 3D Floating Cursor has a 3D coordinate assoa result, wherever you move the 3D Floating Curcoordinate is displayed. To ensure the accuracy ocollected in Stereo Analyst for ArcGIS, the 3D Fpositioned so that it rests on the feature being co3D Floating Cursor rests on the ground or feature3D coordinate is computed and then the feature ccollected. (See “Adjusting the position of the 3Don page 133 for information about placing the 3Don the feature of interest.)

INTRO 5

Acc

SteredatasrasteLANCata

Newto thalso Persobe m

Edi

Usincan ein StechooAttri

acy.

DUCING STEREO ANALYST FOR ARCGIS

You can easily update feature vertices to greater accur

essing image and feature data

o Analyst for ArcGIS supports all of the raster and feature et formats currently supported by ArcMap. For example, r format support includes: ArcSDE rasters, ERDAS® 7.5 , ERDAS IMAGINE® (.img), ERDAS Raw, ESRI Image logs, GRID, GRID Stack, and Windows BMP.

feature datasets can be created using ArcCatalog. In addition is, existing CAD formats such as DGN, DWG, and DXF are supported by ArcMap for use in Stereo Analyst for ArcGIS. nal and multiuser geodatabases created using ArcCatalog can

aintained to include up-to-date feature data.

t ing features

g the editing tools you’re already familiar with in ArcMap, you dit features that you create and features that you have imported reo Analyst for ArcGIS. Simply display the Editor toolbar and

se the edit tools you need to modify feature datasets. bution information is updated accordingly.

LYST FOR ARCGIS

Updating feature data

rded by an image. ages” on page 70

D, but are also

e.

6

Using SHighlyfor infoaccurat

In the pi

USING STEREO ANA

tereo Analyst for ArcGIS, feature datasets can be updated to reflect the geography on the earth’s surface as reco accurate oriented images are used as a reference source for updating feature datasets. (See “Creating oriented imrmation about oriented imagery.) Using Stereo Analyst for ArcGIS, updates to feature datasets are not only in 3e 2D (planimetric) updates.

cture on the left, before editing, this road is not positioned correctly. In the picture on the right, the road clearly follows the featur

7

Collecting features

t new features. To feature of interest

dow.

INTROD

Using tensure being c

The pict

UCING STEREO ANALYST FOR ARCGIS

he Editor tools you’re probably already familiar with in ArcMap, Stereo Analyst for ArcGIS allows you to collecthe accuracy of the GIS data collected using Stereo Analyst for ArcGIS, the 3D Floating Cursor must rest on the ollected.

ure on the left shows features collected in ArcMap; the picture on the right shows the same features collected in the Stereo Win

LYST FOR ARCGIS

Collecting points for the creation of elevation models

a digital elevation accurate 3D point t™, or ERDAS

s collected in Stereo

8

Stereo model (featureIMAGI

The pictAnalyst

USING STEREO ANA

Analyst for ArcGIS allows you to collect elevation information directly from oriented imagery without requiring DEM). Since an accurate 3D digital representation of the earth’s surface is created on the screen using imagery, datasets can be collected. With mass point data, you can easily create DEMs using Spatial Analyst™, 3D AnalysNE.

ure on the left shows mass points collected in Stereo Analyst for ArcGIS. The DTM on the right was generated from the mass pointfor ArcGIS. ERDAS IMAGINE was used to create the DTM.

9

Using ArcMap and Stereo Analyst for ArcGIS

include the Stereo lbar and provides ata in the Stereo

d.

INTROD

Stereo Analysaccess Windo

The Ste


Analyst for ArcGIS adds three toolbars and several new features to ArcMap when it is installed. The three toolbars t toolbar, the Stereo View toolbar, and the Stereo Enhancement toolbar. The Stereo Analyst toolbar is the main tooto importers and exporters as well as preference settings. The Stereo View toolbar provides tools to manipulate dw. The Stereo Enhancement toolbar controls the operation of image enhancement in the Stereo Window.

reo Window, which is the middle window in the above picture, can be docked inside the ArcMap application, or can be undocke

LYST FOR ARCGIS10

Usinimagthe S

You the AArcMcolleWind

StereArcMlayerAnalthe foinfor

Wh

As ycreatalso ArcCGIS

USING STEREO ANA

mapping project.

g these tools, you can perform actions such as selecting an e pair from a graphic display and modifying the behavior of tereo Window.

also have the option of embedding the Stereo Window within rcMap window. The Stereo Window is an extension to the ap window and is used as the workspace for updating and

cting new feature datasets. Work conducted in the Stereo ow is simultaneously reflected in the ArcMap data view.

o Analyst for ArcGIS adds additional functionality to ap. This includes a new Stereo Analyst for ArcGIS footprint

and 3D snap options for feature editing. (See “Using Stereo yst for ArcGIS with ArcMap” on page 119 for information on otprint layer and “Using 3D Snap” on page 162 for

mation about the 3D snap options.)

at can you do with Arc Catalog?

ou know, ArcCatalog is an application with which you can e and manage GIS data. ArcCatalog’s ability to create layers is applicable to Stereo Analyst for ArcGIS functionality. Use atalog to manage the feature datasets that often accompany a

11

Learning about Stereo Analyst for ArcGISIS &

nical support, see eived with Stereo

er Support at stems on the Web

refer to “Getting Getting more al Support is Web at

ng

g about Stereo the Web site e Training link to Registration.

to GISs, GIS ong instructor-led books to find

and pocketbook. esri.com/

INTROD

If you a(GISs),ArcMathese amuch e

If you’works, learn hnavigatdimensversa, cis writtthe exarather, functio

Find

This boupdatinthat yoare intryou wayou hav

Getti

Useful Consulhow to


nt. A glossary is provided to help you understand any terms en’t seen before.

ng help on your computer

information can be found in the On-line Help system. t it to learn how to use Stereo Analyst for ArcGIS. To learn use Help, see the Using ArcMap book.

Training Centers, Course Schedules, and Course

ESRI education solutions

ESRI provides educational opportunities related applications, and technology. You can choose amcourses, Web-based courses, and self-study workeducational solutions that fit your learning style For more information, visit the Web site <www.education>.

re just learning about geographic information systems you may want to read the books about ArcCatalog and p: Using ArcCatalog and Using ArcMap. Knowing about pplications makes your use of Stereo Analyst for ArcGIS asier.

re ready to learn about how Stereo Analyst for ArcGIS refer to chapter 2, “Quick-start tutorial”. In chapter 2 you’ll ow to change image pairs’ display using enhancement tools, e in a 3D digital space in the Stereo Window, create three-ional (3D) feature datasets from 2D feature datasets and vice ollect feature datasets, and edit feature datasets. The tutorial en so that you can do the exercises using your computer and mple data supplied with Stereo Analyst for ArcGIS. If you’d you can simply read the tutorial to learn about the nality of Stereo Analyst for ArcGIS.

ing answers to questions

ok describes the typical workflow involved in creating and g GIS data for mapping projects. The chapters are set up so

u first learn the theory behind certain applications, then you oduced to the typical workflow you’d apply to get the results

Contacting Leica Geosystems GMapping

If you need to contact Leica Geosystems for techthe product registration and support card you recAnalyst for ArcGIS. You can also contact Custom1-404-248-9777. You may also visit Leica Geosyat <www.gis.leica-geosystems.com>.

Contacting ESRI

If you need to contact ESRI for technical supporttechnical support” in the On-line Help system’s “help” section. The telephone number for Technic1-909-793-3774. You may also visit ESRI on the<www.esri.com>.

Leica Geosystems GIS & MappiEducation Solutions

Leica Geosystems offers instructor-based traininAnalyst for ArcGIS. For more information, go to<www.gis.leica-geosystems.com> and follow th

LYST FOR ARCGIS
USING STEREO ANA12

13

22Qui

ereo Analyst for

you’ll learn how .

about populating imagery.

page 31, you’ll dataset. You’ll ing an elevation

n how to use

u’ll learn how to

nd ArcGIS ial data that location of the y.

IN THIS C

• ExerciseAnalyst

• Exerciseimages

• Exercise3D to 2D

• Exercise3D

• Exercisefeatures

ck-start tutorial

This chapter provides the hands-on training you’ll need to use StArcGIS to complete your own mapping projects.

In “Exercise 1: Starting Stereo Analyst for ArcGIS” on page 14, to set up the work environment to run Stereo Analyst for ArcGIS

In “Exercise 2: Adding oriented images” on page 18, you’ll learn the Stereo Window with raster data and how to best examine the

In “Exercise 3: Converting features—3D to 2D and 2D to 3D” onlearn about how to create a 2D feature dataset from a 3D feature also learn how to create a 3D feature dataset from a 2D dataset ussource.

In “Exercise 4: Collecting features in 3D” on page 42, you’ll learediting tools to collect 3D feature data in the Stereo Window.

Finally, in “Exercise 5: Editing existing features” on page 58, youpdate existing features.

To start this tutorial, you must have Stereo Analyst for ArcGIS ainstalled on your system. Also, you must have access to the tutoraccompanies the installation CD. Ask your administrator for the tutorial data if you can’t find it in the default installation director

HAPTER

1: Starting Stereo for ArcGIS

2: Adding oriented

3: Converting features— and 2D to 3D

4: Collecting features in

5: Editing existing

LYST FOR ARCGIS

Exercise 1: Start ing Stereo Analyst for ArcGIS

rcGIS

tion to create a

enu, then click

2

1

USING STEREO ANA14

In the following exercises, we’ve assumed that you are using a single-monitor workstation that is configured for use with ArcMap and Stereo Analyst for ArcGIS. If you have a dual-monitor configuration, you may spread out the applications so that the ArcMap application is displayed on one monitor and the Stereo Window and the Stereo Analyst for ArcGIS toolbars are displayed on the other monitor. This type of setup is ideal for productive feature collection.In this scenario, the ArcMap display serves as the cartographic station for verifying features that have been collected or edited, and the Stereo Window display serves as the main focus for collecting and editing feature datasets.In this exercise, you’ll learn how to start Stereo Analyst for ArcGIS and display all of the toolbars associated with Stereo Analyst for ArcGIS. You’ll be able to use the toolbars to gain access to all of the key functionality in Stereo Analyst for ArcGIS.

Preparing

This exercise assumes that you have already successfully completed installing Stereo Analyst for ArcGIS on your computer. If you haven’t installed Stereo Analyst for ArcGIS, do so now.

Star ting ArcMap

1. Click the Start button on your desktop, then point to Programs, then point to ArcGIS.

2. Click ArcMap to start the application.

Adding the Stereo Analyst for Aextension

1. If the ArcMap dialog opens, keep the opnew map, then click OK.

2. In the ArcMap window, click the Tools mExtensions.

1

15

een selected,

bars, then click ArcMap

4

QUICK-START TUTORIAL

3. On the Extensions dialog, click the check box for Stereo Analyst to add the extension to ArcMap.

Once the Stereo Analyst check box has bthe extension is activated.

4. Click Close on the Extensions dialog.

Adding toolbars

1. Click the View menu, then point to ToolStereo Analyst to add that toolbar to the window.

2

3

LYST FOR ARCGIS

With the Stereo View toolbar, you can enable many of odes, such as ursor Mode.

s Zoom Mode xtent of the . ize the ArcMap dinates to drive right images,

bars, then click to the ArcMap

u can change h the individual

3

USING STEREO ANA16

With the Stereo Analyst toolbar, you can choose image pairs to view, open the Stereo Window, import and export data, and much more.

2. Click the View menu, then point to Toolbars, then click Stereo View to add that toolbar to the ArcMap window.

the special Stereo Analyst for ArcGIS mthe Terrain Following Mode and Fixed CAdditionally, you can use the Continuouand the image Roam Tool to adjust the eimage pair display in the Stereo WindowOther tools allow the ability to synchronand Stereo Window displays, input coorto a specific location, reverse the left andand update the feature display.

3. Click the View menu, then point to ToolStereo Enhancement to add that toolbar window.

With the Stereo Enhancement toolbar, yothe contrast and brightness display of bot

1

2

17

images that make up the image pair as well as whole


image pairs in the Stereo Window.

4. Arrange the toolbars in the ArcMap window so that you can easily access each of them.

Closing the applications

If you plan to continue on to “Exercise 2: Adding oriented images”, then proceed to “Adding image pairs” on page 18.Otherwise, you may exit ArcMap, which also closes Stereo Analyst for ArcGIS.

What’s next?

In “Exercise 2: Adding oriented images”, you’ll add oriented images to ArcMap and learn how to use some of the Stereo Analyst for ArcGIS tools in the Stereo Window.

LYST FOR ARCGIS

Exercise 2: Adding or iented imagesolder called

rip1_1.img and r images in the

ap data view.

orded with a hotography canned. Each imately 16.5 earth’s surface.

3

4

2

USING STEREO ANA18

In this exercise, you’ll learn how to add multiple oriented images (rasters) to ArcMap and the Stereo Window. The oriented images you’ll be using were created by importing data from the SOCET SET® digital photogrammetry software product.

Preparing

If you’re continuing the tutorial from “Exercise 1: Starting Stereo Analyst for ArcGIS”, you may proceed to the section named “Adding image pairs” below.If you’re starting this exercise from scratch, you should have ArcMap running with an empty data view, and Stereo Analyst for ArcGIS should also be loaded and running. You should also have the three Stereo Analyst for ArcGIS toolbars displayed. Proceed to “Adding image pairs” below.

Adding image pairs

1. Click the Add Data button to select the rasters to add to ArcMap.

2. In the Add Data dialog, navigate to the f\ArcTutor\StereoAnalyst\Images.

3. Shift-click to select the images named ststrip2_2.img. This selects all of the rastelist.

4. Click Add to add the images to the ArcM

The photographs you’re adding were recspecially-fitted camera used to capture pfrom aircraft. Then, they were digitally spixel represents 0.16458 meters (approxcentimeters by 16.5 centimeters) on the

Using over lapping, GIS-ready images

In order to view or update existing features or collect new features, at least two overlapping GIS-ready rasters must be added to ArcMap. See chapter 3 “Working with oriented images” on page 69 to learn more about GIS-ready images.

1

19

Creating pyramid layers

ut the absence

a projection, as

ter dataset, they eleted. Pyramid on and are ters with which


A pyramid layer is a version of the original raster that has been resampled at a lower resolution. A raster dataset commonly has eight pyramid layers. Pyramid layers make navigation and display of large raster datasets much faster at any resolution. Therefore, you should build pyramid layers for each of the raster images in this dataset.1. Click OK on the Create pyramids dialog to build

pyramid layers for the first raster image, strip1_1.img.

A progress meter displays to show the status of the pyramid layer generation.

2. Click OK on the next three Create pyramids dialogs to generate pyramid layers for all of the images you’ll be using in this exercise.

3. Click OK on the dialog alerting you aboof spatial reference information.

You get this message if the images lack in the case of these images.

1

Creat ing pyramid layers

Once pyramid layers have been created for a rasdo not need to be created again unless they are dlayers are contained in files with an .rrd extensitypically located in the same directory as the rasthey are associated.

3

LYST FOR ARCGIS

Once you’ve added the rasters, you can see that the data otography, and ographs. The outh-east to pear diagonal

ir bounding round values in ArcMap.

that the view. To make ayer Properties

lue is 0 (which d displays in se areas appear

the Table of

as an image is what is most ing portion of rth’s surface

USING STEREO ANA20

Once the four raster images are added to ArcMap, they display with each raster footprint (red) and overlap area (yellow) indicated by graphical borders.

view shows two overlapping strips of pheach strip contains two overlapping photaircraft that recorded the imagery flew snorth-west, and as a result the images apwithin ArcMap.

Because the images are diagonal, but theboxes are squares, they have dark backgthat may be distracting as you view themYou’ll fix that next.

Changing properties

You can alter the properties of each image sobackground values are transparent in the datathe change, make a simple adjustment to the Lfor each raster image.In these example images, the background vayou can determine using the Identify tool) anblack. If you set the 0 value to No Color, thotransparent in the data view. 1. Right-click the title of the first image in

contents, then select Properties.

Changing footpr int and over lap display

By default, the display color for image footprints is red, and the display color for overlap areas is yellow. If you would like to use different colors, you can change the defaults on the Stereo Analyst Options dialog.

To access the options, on the Stereo Analyst toolbar click the Stereo Analyst dropdown list, then select Options. On the ArcMap Display tab, use the dropdown lists to access and apply different colors for the Oriented image footprint color and the Selected Image Pair highlight color. Then click OK to accept the changes.

Viewing over lapping areas

Two adjacent rasters that overlap are referred topair. The overlapping portion of the two rastersuseful in the Stereo Window. It is this overlapptwo raster images that is used to recreate the eadigitally and in 3D within the Stereo Window.

21

4. Click OK and the image redisplays in ArcMap with the arently.nging es listed in the

footprints, and

he display in eo Window. To same ArcMap nalyst Options

Stereo Analyst


2. On the Symbology tab of the Layer Properties dialog, click the check box for Display Background Value. Make sure the value is set to 0 (zero).

3. Make sure that the Display Background Value as color block is set to No Color.

background value 0 set to display transp5. Repeat step 1 through step 4 under “Cha

properties” for the remaining three imagTable of contents.Now, only the raster images, their vectoroverlap boundaries display in ArcMap.

Changing the ArcMap display

It is often helpful to have the orientation of tArcMap match that of the display in the Sterensure that this is always the case within thesession, you can set an option on the Stereo Adialog.1. On the Stereo Analyst toolbar, click the

dropdown list, then click Options.

1

42 3

LYST FOR ARCGIS

and close the

anges. You’ll ages’ display in “Working with

USING STEREO ANA22

2. Click the ArcMap Display tab of the Stereo Analyst Options dialog.

3. Click the check box for Orient ArcMap document to Image Pair when Image Pair changes.

4. Click OK to apply the orientation settingStereo Analyst Options dialog.The display of the images in ArcMap chsee how the change compliments the imthe Stereo Window in the section called the Stereo Window” on page 24.

1

23

4

23

ition to indicate age pair

air and is used the earth’s

aphically select g/strip2_2.img. ctive.


Changing the image pair

One or more image pairs consisting of two overlapping raster images may display in ArcMap. When you need to change the geographic extent displayed within the Stereo Window, you can easily select a different image pair in ArcMap.You can change an active image pair whose overlap portion is highlighted in yellow in two ways. Either use the Image Pairs dropdown list or use the Image Pair Selection Tool (on the Stereo Analyst toolbar) to choose the image pair you want to work with.1. On the Stereo Analyst toolbar, click the Image Pairs

dropdown list and select a different pair to make active, strip1_1.img/strip1_2.img.

The yellow overlap graphic changes posthe new active area of overlap for the imstrip1_1.img/strip1_2.img. The new image pair is the active image pto recreate a 3D digital representation ofsurface in the Stereo Window.

2. Click the Image Pair Selection Tool to grthe original area of overlap—strip2_1.imThe button appears recessed when it is a

1

LYST FOR ARCGIS

displays to ecific area of ou’ll learn how

Stereo Window

include the be docked ow.

s card that or ArcGIS uses isplay the 3D

this case, you ge pairs in 3D.

cs/specs.html>

1

USING STEREO ANA24

3. Move your cursor into the data view and position it over the original area of overlap. Notice that the area becomes highlighted in yellow as you move your cursor inside the overlap area.

4. Click to select the overlap area of strip2_1.img/strip2_2.img and make it the active image pair.

5. To disable the Image Pair Selection Tool, click the ArcMap Select Elements button.

Working with the Stereo Window

The Stereo Window is where the image pairrecreate a 3D digital representation of the spinterest as recorded in the oriented images. Yto open it next.

1. On the Stereo Analyst toolbar, click the button.

The ArcMap window contents adjust to Stereo Window. The Stereo Window caneither inside or outside the ArcMap wind

2

3

5

Using graphics cards

If your computer system doesn’t have a graphicsupports quad-buffered stereo, Stereo Analyst fanaglyph rendering techniques to recreate and ddigital representation of the area of interest. In need red/blue anaglyph glasses to view the ima

See the Web site <http://support.erdas.com/spefor a list of supported graphics cards.

25

ed Cursor

3D Floating r of the Stereo neath” it.

ated on the en move it atures overlap ollowing


Initially, the Stereo Window opens in 3-Pane View: one pane for the primary stereo scene, which displays the active image pair; and one pane each for the left and right oriented images of the image pair (the 2-Pane View).

2. Undock the Stereo Window from the ArcMap display by clicking on the window’s title bar and dragging it outside the ArcMap display.

3. Resize the Stereo Window to your liking. The cursor becomes a double-headed arrow at the extents of the window indicating that you can click, hold, and drag the window to a new size.

Adjusting the display

1. Put on your stereo or anaglyph glasses.

You’ll notice that the Stereo Window display is 3D.

2. On the Stereo View toolbar, click the FixMode button.

When Fixed Cursor Mode is active, the Cursor maintains its position in the centeWindow while the image pair moves “be

3. Click and hold on the Z thumb wheel locright-hand side of the Stereo Window, thslightly up and/or down until the same fein both the right and left images. In the f

2

LYST FOR ARCGIS

picture, the images have been adjusted so that the area 4. Click the 1-Pane View button located at the bottom of View from the

in to exit that

de, the button View toolbar.

see the entire Stereo Window.

USING STEREO ANA26

around the 3D Floating Cursor, which is circled in red, is set for optimal stereo viewing.

The cursor in the Stereo Window is called the 3D Floating Cursor because it can float on, below, or above a feature.By adjusting the Z thumb wheel, the height of the 3D Floating Cursor is modified via the movement of the images of the image pair. You can see the elevation change in the coordinates of the location of the 3D Floating Cursor, which are displayed in the lower-left status bar located within the Stereo Window.

the Stereo Window to remove the 2-PaneStereo Window configuration.This enlarges the area of stereo display.

5. Click the Fixed Cursor Mode button agamode. When you have exited Fixed Cursor Mono longer appears recessed on the Stereo

Adjusting the zoom ratio

1. Click the Zoom to Data Extent button toextent of the image pair displayed in the

3

4

27

2. On the Stereo View toolbar, click the Zoom In By 2 omfortably see in 3D within

your cursor Window and e.

Roam Mode. just the image


If you are viewing in anaglyph mode, the left image of the image pair appears red for nonoverlap areas; the right image appears blue for nonoverlap areas; and the overlap area is grey. If you are viewing in quad-buffered stereo, the entire area appears grey, but you can see in 3D only in the overlap area.

button a number of times until you can cfeatures on the earth’s surface displayedthe Stereo Window.

3. Click the Roam Tool button, then move (which appears as a hand) into the Stereodouble-click to activate Auto Roam Mod

4. The hand changes into an arrow in AutoMove your mouse in any direction to adpair’s position in the Stereo Window.

1

23

LYST FOR ARCGIS

Adjusting brightness and contrast

f the image pair viewing and

r; however, you ging the setting

e sure that the s.

k, hold, and and left to see in the Stereo

air with

USING STEREO ANA28

The arrow’s proximity to the center of the Stereo Window determines the speed of the Auto Roam Mode. If you are close to the center of the Stereo Window, the speed is slow; if you are close to the edges of the Stereo Window, the speed is fast.

5. Once you find an area that interests you, double-click in the Stereo Window again to return to normal Roam Mode.Notice that the 3D Floating Cursor recenters in the middle of the Stereo Window when you exit Auto Roam Mode.

You can adjust the brightness and contrast odisplayed in the Stereo Window to suit yourfeature collection needs. By default, both images are adjusted togethecan also adjust each image separately by chanin the Adjust dropdown list.1. On the Stereo Enhancement toolbar, mak

Adjust dropdown list shows Both Image

2. On the Stereo Enhancement toolbar, clicmove the Brightness thumb wheel right the changes in the image pair displayed Window.

The following picture shows the image pdecreased brightness.

4

1

2

29

air with


3. Click the Reset Brightness button.

4. On the Stereo Enhancement toolbar, click, hold, and move the Contrast thumb wheel right and left to see the changes in the image pair displayed in the Stereo Window.

The following picture shows the image pincreased contrast.

5. Click the Reset Contrast button.

3

4

5

LYST FOR ARCGIS

6. Close the Stereo Window by clicking the Close button in

USING STEREO ANA30

the top right corner of the window.


If you plan to continue on to “Exercise 3: Converting features—3D to 2D and 2D to 3D”, do not remove any of the images currently displayed in ArcMap.Otherwise, you may Exit ArcMap, which also closes Stereo Analyst for ArcGIS.

What’s next?

In “Exercise 3: Converting features—3D to 2D and 2D to 3D”, you’ll learn how to use Stereo Analyst for ArcGIS tools to convert your 3D features to 2D features and vice versa.

6

31

Exercise 3: Convert ing features—3D to 2D and 2D to 3Dting you about ion.n ArcMap es” on page 20 ges’ display.

atabase that D. Stereo

ity that quickly how to use that

Stereo Analyst eatures to 2D.

. In this dialog, the parameters o 2D features.lick the Open

1


Two options are available for converting feature datasets—one is actual and one is virtual. The actual option converts 3D features to 2D features, or converts 2D features to 3D features. In either case, the result is a new, separate file.The virtual option converts a 2D feature dataset to 3D, but a separate file is not created. For more information on Virtual 2D To 3D, please refer to “Using Virtual 2D To 3D” on page 89.In this exercise, first you’ll learn how to convert 3D features to 2D. Then, you’ll learn how to convert 2D features to 3D using an elevation source.

Preparing

If you’re continuing the tutorial from “Exercise 2: Adding oriented images”, you may progress to the section named “Converting 3D features to 2D” on page 31.If you’re starting this exercise from scratch, you should have ArcMap and Stereo Analyst for ArcGIS running with an empty data view. You should have the Stereo Analyst and Stereo View toolbars displayed. Proceed to “Adding image pairs” below.

Adding image pairs

1. Click the Add Data button to select the rasters to add to ArcMap.

2. In the Add Data dialog, navigate to the folder called \ArcTutor\StereoAnalyst\Images.

3. Shift-click to select the images named strip1_1.img and strip2_2.img. This selects all of the images in the list.

4. Click Add to add the images to ArcMap.

5. If necessary, click OK on the dialog alerthe absence of spatial reference informatIf the display of the background values ibothers you, refer to “Changing propertifor instructions about how to fix the ima

Converting 3D features to 2D

The features, which are in the example geodcomes with Stereo Analyst for ArcGIS, are 3Analyst for ArcGIS provides you with a utilconverts 3D feature data to 2D. You’ll learnutility in the next series of steps.1. On the Stereo Analyst toolbar, click the

dropdown list and choose Convert 3D F

The Convert Features to 2D dialog opensyou can define input options and modifyassociated with converting 3D features t

2. On the Convert Features to 2D dialog, cbutton.

LYST FOR ARCGIS

5. Click Open on the Input Features dialog to add the Convert

dow, then nd click to R_INDEX.ick to select the

AILROAD,

ing features

tomatically

aset, but with t, the contents

1

2

USING STEREO ANA32

3. In the Input Features dialog, navigate to the folder named \ArcTutor\StereoAnalyst\Geodatabase.

4. Click to select the geodatabase file named sampleAltdorfFME.mdb.

feature classes in the geodatabase to the Features to 2D dialog.

Select ing feature c lasses to make 2D

1. Scroll to the top of the Select classes winposition your cursor inside the window aselect the feature class named CONTOU

2. Hold the Ctrl key on the keyboard and clclasses: CONTOUR_INTERMEDIATE,FREEWAY, HOUSE, PAVED_ROAD, RRIVER, and SPOTHEIGHT.

Choosing other set t ings and convert

1. Notice that the Output dataset name is auentered for you. By default, it is named like the input datthe additional element “_2D”. As a resul

2

4

5

3

33

of the original dataset are not changed, and a new file is 4. When the process is complete, indicated by the status he ArcMap ures to 2D

r the oriented

buttons to

get information


generated. Unless you select a different location, the new file is placed in the same location as the input dataset.

2. Notice that the Add converted 2D feature classes to ArcMap document check box is selected by default.This option adds the converted feature datasets to the ArcMap Table of contents and data view immediately after conversion. If this box is not checked, then you may add the 2D dataset to ArcMap at a later time like any other file.

3. Click Convert on the Convert Features to 2D dialog to begin the conversion process.A status bar displays at the bottom of the ArcMap window. You can view the percentage complete there as the process runs.

bar’s closure in the lower-left corner of twindow, click Done on the Convert Featdialog to close it.The new, 3D feature classes display oveimages in ArcMap.

5. Use the ArcMap Fixed Zoom In and Panclearly view the features in ArcMap.

You can use the ArcMap Identify tool to about individual features.

1

2

3 4

5

LYST FOR ARCGIS

Conf i rming features are 3D

tions dialog.

ust created, u can convert it

on source.Stereo Analyst .

2

USING STEREO ANA34

You can use the Virtual 2D To 3D tab of the Stereo Analyst Options dialog to determine whether or not features are 2D. 3D features are not eligible for use in Virtual 2D To 3D and will not display on the Virtual 2D To 3D tab of the Stereo Analyst Options dialog. Since this tool is only operational with 2D features; therefore, you’ll use it in this case to confirm that the features were converted to 2D. For more information, see “Using Virtual 2D To 3D” on page 89 in chapter 4 “Working with 3D data”.1. From the Stereo Analyst toolbar, click the Stereo

Analyst dropdown list and choose Options.

2. On the Stereo Analyst Options dialog, click the Virtual 2D To 3D tab.

3. Notice that all of the features you chose for 3D to 2D conversion in the previous section of this exercise are listed in the Selected features window of the Virtual 2D To 3D tab.

4. Click OK to close the Stereo Analyst Op


Next, you’ll use the 2D feature dataset you jsampleAltdorfFME_2D.mdb, and see how yoback to a 3D feature dataset using an elevati1. On the Stereo Analyst toolbar, click the

dropdown list and choose Features to 3D

13 4

35

3. In the Input Features dialog, navigate to the folder base.reated in the rting 3D .mdb.

to add the Convert

sses associated in the Select log.o convert from exercise are

EX, AY, HOUSE,

and

4

5

3


The Convert Features to 3D dialog opens. In this dialog, you can define input options and modify the parameters associated with converting 2D features to 3D features.

2. On the Convert Features to 3D dialog, click the Open button.

named \ArcTutor\StereoAnalyst\Geodata4. Click to select the geodatabase file you c

previous section of this exercise, “Convefeatures to 2D”, sampleAltdorfFME_2D

5. Click Open on the Input Features dialogfeature classes in the geodatabase to the Features to 3D dialog.

Select ing feature c lasses to make 3D

By selecting a geodatabase, all the feature clawith the geodatabase are automatically listedclasses list of the Convert Features to 3D dia1. Notice that all of the classes you chose t

3D to 2D in the previous section of this listed in the Select classes window. Those classes include: CONTOUR_INDCONTOUR_INTERMEDIATE, FREEWPAVED_ROAD, RAILROAD, RIVER, SPOTHEIGHT.

1

2

LYST FOR ARCGIS

2. Make sure that all of the classes are selected

gate to the folder.

2

4

5

3

USING STEREO ANA36

(highlighted) in the list.

Select ing an e levat ion source

The elevation source is where Stereo Analyst for ArcGIS gets the elevation information for the 2D features in order to convert them to 3D features.

1. Under Elevation Source, click the Raster or TIN surface option.

2. Click the Open button for the Raster or TIN surface option.

3. In the Raster or Tin Surface dialog, navidirectory \ArcTutor\StereoAnalyst\DEM

4. Select the file Altdorf_1m_dem.img.

1

2 1

37

5. Click Open on the Raster or Tin Surface dialog to accept

by the status he ArcMap ures to 3D

r the oriented created) in

buttons to

1

4


the DEM raster file as the elevation source.

Choosing other set t ings and convert ing features

1. Notice that the Output dataset name is automatically entered for you. By default, it is named like the input dataset, but with the additional element “_3D”. As a result, the contents of the original dataset are not changed, and a new file is generated.

2. Notice that the Add converted 3D feature classes to ArcMap document check box is selected by default.This option adds the converted feature datasets to the ArcMap Table of contents and data view immediately after conversion. If this box is not checked, then you may add the 3D dataset to ArcMap at a later time like any other file.

3. Click Convert on the Convert Features to 3D dialog to begin the conversion process.A status bar displays at the bottom of the ArcMap window. You can view the percentage complete there as the process runs.

4. When the process is complete, indicatedbar’s closure in the lower-left corner of twindow, click Done on the Convert Featdialog to close it.The new, 3D feature classes display oveimages (and 2D features you’ve already ArcMap.

5. Use the ArcMap Fixed Zoom In and Panclearly view the features in ArcMap.

Ensur ing accuracy

The accuracy of the elevation source used to convert a feature dataset to 3D impacts the quality and the amount of editing required to ensure that the features reflect what is on the earth’s surface. The more an elevation source reflects the topography on the ground, the less feature editing is required.

3

2

5

LYST FOR ARCGIS

You might want to use the Source tab of the Table of Viewing 3D features in the Stereo Window

ature datasets , you can view rea, which is

Stereo Window

ewing glasses y adjusts to displays the reo Window in ing is retained. om Out By 2 ter and feature

ert Features to ew and the

1

USING STEREO ANA38

contents to confirm which feature classes are the 2D feature classes, and which are the 3D feature classes. Now that you’ve successfully updated the fe

using a raster DEM as your elevation sourcethe features in the 3D representation of the acreated using an image pair.1. On the Stereo Analyst toolbar, click the

button to open a Stereo Window.

Make sure you put on the appropriate vi(anaglyph or stereo). The ArcMap displaaccommodate the Stereo Window, which3D feature data. If you undocked the Stethe previous exercise, that undocked sett

2. On the Stereo View toolbar, click the Zobutton until you see the extent of the rasdata displayed in the Stereo Window.

All of the features you chose in the Conv3D dialog display in the ArcMap data viStereo Window.

Al igning features

The features may not entirely align with the oriented images in ArcMap. This is common since the raw pixels in the oriented images have not been transformed and then projected to create a new raster dataset. This process is commonly referred to as orthorectification.

Viewing the same features and oriented images in the Stereo Window yields better results since Stereo Analyst for ArcGIS resamples raw pixels on the fly. To learn more about this, refer to “Applying epipolar correction” on page 124.

2

39

that allow you w. You’ve with its portions of the

t mouse button ndow to view a

erlap between to maintain the


3. On the Stereo View toolbar, click the Default Zoom button to display the oriented images at a 1 image pixel to a 1 screen pixel resolution. This is referred to as 1 to 1 Zoom.

Using tools to view features

Stereo Analyst for ArcGIS comes with toolsto rapidly view the data in the Stereo Windoalready tried the Roam Tool and are familiarfunctionality, so apply it again to view otherimage pair.

1. Click the Roam button, then hold the lefand drag the image pair in the Stereo Widifferent area.Make sure you stay within the area of ovthe left and right image of the image pair3D view in the Stereo Window.

3

LYST FOR ARCGIS

, click and hold use towards

to adjust the de. nually Toggle

e Zoom In Tool

USING STEREO ANA40

2. Click the Zoom In Tool button and click inside the Stereo Window until you see a feature of interest.You can also draw a rectangle in the Stereo Window with the Zoom In Tool to select an area to view.

When Zoom In Mode is applied, the feature that you clicked on is automatically repositioned so that it displays in the center of the Stereo Window. You may continue to click inside the Stereo Window until the feature is displayed at the resolution you want, which is indicated in the Scale window of the Stereo View toolbar.

3. If necessary, use the Z thumb wheel in conjunction with Fixed Cursor Mode to align the left and right images so that features overlap properly. Once they are aligned, make sure that Fixed Cursor Mode is turned off.

4. With the Zoom In Tool button still activethe mouse scroll wheel, then drag the moand away from you. This is another waydisplay, called the Continuous Zoom Mo

5. On the Stereo View toolbar, click the Ma3D Floating Cursor button to deselect thbutton.

Enter ing and exi t ing Auto Roam Mode

Remember, you can double-click with the left mouse button when in Roam Mode to enable Auto Roam Mode. Double-click again to return to normal Roam Mode.

12

5

41

6. On the Stereo View toolbar, click the Default Zoom Closing the applications

ting features in therwise, you . This closes

layed in the ck and choose

g and tents, then

2.img remains

ols used to Analyst for

ures.


button to display the oriented images at a 1 image pixel to a 1 screen pixel resolution.

The image pair display in the Stereo Window updates.

If you plan to proceed to “Exercise 4: Collec3D”, then perform the next series of steps. Omay exit the ArcMap application at this timeStereo Analyst for ArcGIS as well.1. Shift-click all of the feature classes disp

ArcMap Table of contents, then right-cliRemove.

2. Shift-click the images named strip1_1.imstrip1_2.img in the ArcMap Table of conright-click and choose Remove.

Only the image pair strip2_1.img/strip2_in the Table of contents.

What’s next?

Now that you’re familiar with some of the tomanipulate the image pair’s display in StereoArcGIS, you’re ready to start collecting feat

6

LYST FOR ARCGIS

Exercise 4: Col lect ing features in 3Dn ArcMap es” on page 20 ges’ display.

access the

Stereo Analyst

Map, you can ArcMap Table main displayed.

tions. Click the ns. Click the otprints of

1

2

USING STEREO ANA42

To collect features in Stereo Analyst for ArcGIS, you use tools in the ArcMap Editor you’re probably already familiar with. You can use these tools to collect features in the Stereo Window—the difference, of course, is that you’re collecting features in 3D.Collecting features in 3D can be made easier by using the 2-Pane View of the Stereo Window. This method is described in detail in this exercise.

Preparing

If you’re continuing the tutorial from “Exercise 3: Converting features—3D to 2D and 2D to 3D”, proceed to the section named “Accessing device settings” on page 42.If you’re starting from scratch, you should have both ArcMap and Stereo Analyst for ArcGIS running on your machine. You should have an empty data view and Stereo Window, and the Stereo Analyst and Stereo View toolbars displayed. Proceed to “Adding images” below.

Adding images

1. Click the Add Data button to select the rasters.2. In the Add Data dialog, navigate to the folder named

\ArcTutor\StereoAnalyst\Images.3. Ctrl-click to select the images named strip2_1.img and

strip2_2.img.4. Click Add to load the rasters in the Stereo Window and

ArcMap.5. If necessary, click OK on the dialog alerting you about

the absence of spatial reference information.

If the display of the background values ibothers you, refer to “Changing propertifor instructions about how to fix the ima

Accessing device sett ings

To configure the system mouse, you need toDevices dialog.1. On the Stereo Analyst toolbar, click the

dropdown list.2. Click the Devices option.

Enhancing per formance

If you want to enhance the performance of Arcturn off the display of the oriented images in theof contents. The footprint and overlap extents re

You can also do this via the Stereo Analyst OpStereo Analyst dropdown list and choose OptioArcMap Display tab, then click to select the Fooriented images option, then click OK.

43

Configuring the system mouse This opens the System Mouse Properties dialog where can be

is-To-Ground. control the and Z

acceptable for ses the 3D

in ground space ent of the X, Y, and Z

lue for the hich is 1.00 by

window that

scroll wheel map unit ating Cursor. s are


Before collecting new features, it’s important to configure your input device for optimal use. An input device is the computer system’s peripheral device that is used to control the 3D Floating Cursor in Stereo Analyst for ArcGIS. In this exercise, you’ll be using the system mouse.

Sett ing propert ies

1. On the Devices dialog, click to highlight System Mouse from the Device Selection list.

2. Click Properties on the Devices dialog.

settings that control mouse performancechanged.

One of the most important settings is AxThe settings in this section of the dialog sensitivity of mouse movement in X, Y, (elevation). Usually, the X and Y settings of 1.00 aredigitizing. Increasing these values increaFloating Cursor’s amount of movement in the Stereo Window relative to movemmouse (or other digitizing device) in thedirections and vice versa. For this exercise, you should alter the vascroll wheel (that affects Z elevation), wdefault.

3. Type the value 0.005 in the ScrollWheelcorresponds to Z.This means that every movement of the (either up or down) results in a +/- 0.005adjustment of the elevation of the 3D FloAll other default system mouse propertieappropriate for this exercise.

Using input (d ig i t iz ing) devices

Possible input devices include: Immersion’s SoftMouse, ITAC Systems’ Mouse-Track Professional, Leica Geosystems’ TopoMouse, and the standard system mouse.

See the Stereo Analyst for ArcGIS On-line Help for more information.

21

LYST FOR ARCGIS

stem Mouse ment by og. You access stem.

log, there is a ed box, below) est results, this

quals 32 points, in the Stereo heel moves the

s.

ent of the scroll 6 units in

USING STEREO ANA44

4. Click OK to return to the Devices dialog.

3

4

Contro l l ing Z movement

In addition to the ScrollWheel setting on the SyProperties dialog, you can also control Z movechanging a setting on the Mouse Properties dialthis dialog through the Control Panel of your sy

On the Buttons tab of the Mouse Properties diasetting for the Scrolling Size (indicated by the rthat affects the scroll wheel of the mouse. For bvalue should be set to 1 line. The 1 line setting ewhich is equivalent to 32 units in ground spaceWindow. Therefore, each “click” of the scroll w3D Floating Cursor up or down 32 ground unit

If the Scrolling Size is set to 3 lines, then movemwheel equals 96 points, which is equivalent to 9ground space, and so on.

45

Mapping buttons 3. Notice that the Currently assigned to window shows

list and check

roll wheel) n as the

ault settings. n easily change

utton Mapping

5


It’s also possible to change the functions of the mouse buttons.1. With the System Mouse still selected on the Devices

dialog, click Button Mappings to open the System Mouse Button Mapping dialog.

The System Mouse Button Mapping dialog has three components. These are: Categories, Commands/Buttons, and Customize Button Assignment. Categories lists the various ArcMap function categories, the Commands/Buttons window shows the commands within each of the Categories, and the Customize Button Assignment section of the dialog is where the current button assignments can be viewed and changed. The Customize Button Assignment area contains a Press/Select device button dropdown list that can be used to display the different mouse button assignments. When a selection is made, the button’s function is listed in the Currently assigned to window.

2. Click the Press/Select device dropdown list and choose Mouse Left Button.

Start Feature Collection.4. Click the Press/Select device dropdown

the settings for the other mouse buttons.The middle mouse button (that is, the scremains unassigned so that it can functioelevation control in the Stereo Window.For this exercise, you’ll be using the defHowever, it is useful to know that you cathe button mappings via this dialog.

5. Click Close to exit the System Mouse Bdialog.

1

2

3

LYST FOR ARCGIS

tereo Analyst

choose a color r the 3D

idth of the 3D xels (points), to

d choose the list, such as

s

r shapes, please on page 136 in

USING STEREO ANA46

6. Click Close on the Devices dialog.

Changing the 3D Floating Cursor

You can easily change the way the 3D Floating Cursor looks in the Stereo Window. This may make features easier to collect.1. On the Stereo Analyst toolbar, click the Stereo Analyst

dropdown list and choose Options.

2. Click the 3D Floating Cursor tab of the SOptions dialog.

3. Click the Cursor color dropdown list andother than white, which is the default, foFloating Cursor.

4. Click the up arrow to increase the Line wFloating Cursor, which is measured in pi6.00.

5. Click the Cursor shape dropdown list ananother 3D Floating Cursor shape from Open X with dot.

6

Restor ing defaul t set t ings

The default settings files for the system mouse and all other supported devices are contained in the following location: \arcexe83\Raster\ButtonMappings\StereoAnalyst. You can also click Reset All on the corresponding Button Mapping dialog to return to the original, default settings.

1

Changing 3D Float ing Cursor shape

For more information about 3D Floating Cursorefer to “Selecting 3D Floating Cursor options”chapter 6 “Applying the 3D Floating Cursor”.

47

When the Stereo Window is in 3-Pane View, it includes reo Window. nd right images igitizing g Cursor rests

ale dropdown

mon techniques dd the feature

ata button.

lder..mdb.

uildings, portation.


6. Click Apply on the 3D Floating Cursor tab of the Stereo Analyst Options dialog.

7. Click OK to close the Stereo Analyst Options dialog.

Setting up the Stereo Window

1. If the Stereo Window is not in 3-Pane View, click the 3-Pane View button at the bottom of the Stereo Window.

the 2-Pane View at the bottom of the SteThe 2-Pane View, which shows the left aof the image pair, is very helpful when dfeatures to make sure that the 3D Floatinon the same feature in each image.

2. On the Stereo View toolbar, click the Sclist and choose 150%.

Adding feature classes

In this exercise, you’ll learn some of the comused in the collection of features. First, you aclasses.1. On the ArcMap toolbar, click the Add D2. In the Add Data dialog, navigate to the

\ArcTutor\StereoAnalyst\Geodatabase fo3. Double-click the file sampleAltdorfFME4. Ctrl-click to select the datasets named B

Ground Points, Hydrography, and Trans

2

7 6

3

5

4

1

2

LYST FOR ARCGIS

Collecting a polygon feature

number of te a shape. For esented by four

ou’ll be coordinates e pair active. Position Tool

l dialog, type

window. log.

3

2

USING STEREO ANA48

The Buildings feature dataset has the layers HOUSE, HOUSE_EXTENSION, and STORAGE_TANKS. The Ground Points feature dataset has the layers FULL_CONTROL_POINT and SPOTHEIGHT. The Hydrography feature dataset has the layers DRAIN, POND, RIVER, STORM_DRAIN, and STREAM. The Transportation feature dataset has the layers FOOTPATH, FREEWAY, PAVED_ROAD, RAILROAD, and TRACK.

5. Click Add on the Add Data dialog.

Displaying the Editor toolbar

The ArcMap editing tools let you collect and edit features quickly in the Stereo Window. If you don’t have the Editor toolbar displayed, click the View menu, then point to Toolbars, then click Editor to display the toolbar.

Polygon features are created by collecting a vertices, which are eventually closed to creaexample, a rectangular building may be reprconnected vertices.Locat ing the polygon feature

The easiest way to locate the first polygon ydigitizing is to use the 3D Position Tool. Theyou’re going to enter correspond to the imagstrip2_1.img/strip2_2.img, so make sure it is1. On the Stereo View toolbar, click the 3D

button.

2. In the X window of the 3D Position Toothe X coordinate 691121.3.

3. Type the Y coordinate 191339.7.You don’t need to enter a value in the Z

4. Click Apply on the 3D Position Tool dia

4

5

3

1

4

49

The position of the 3D Floating Cursor adjusts to reflect

thumb wheel, e roof of the ge and the left rsor Mode

se Start Editing.

Task window

opdown list and

ta view ientation of the n the display in

ay in the Stereo enu on the

the ArcMap Map document lick OK.

the display in o unrotate the


the coordinates you enter in the 3D Position Tool. This is the first building you’ll digitize.

5. On the Stereo View toolbar, click the Synchronize Geographic Displays button so that the view in the ArcMap data view approximates that in the Stereo Window.

Prepar ing to col lect the feature

1. Using the Fixed Cursor Mode and the Z adjust the position of the images until thbuilding feature overlaps in the right imaimage. Make sure you deselect Fixed Cuwhen you are finished.

2. Click the Editor dropdown list and choo

3. On the Editor toolbar, make sure that thedisplays Create New Feature.

4. On the Editor toolbar, click the Target drchoose HOUSE.

5

Orient ing the ArcMap display

While the location displayed in the ArcMap daapproximates that in the Stereo Window, the orimage pair in ArcMap may appear different thathe Stereo Window.

To orient the ArcMap display to match the displWindow, select the Stereo Analyst dropdown mStereo Analyst toolbar, then click Options. On Display tab, click the check box for Orient Arcto Image Pair when Image Pair changes. Then c

Note, however, if you uncheck this check box, ArcMap won’t return to the original rotation. Tdisplay, use the ArcMap Data Frame Tools.

2

LYST FOR ARCGIS

7. Adjust the scroll wheel on the mouse up and down until n the same w (the roof of

may want to the individual hen the

left pane t pane, you are

USING STEREO ANA50

5. On the Editor toolbar, click the Sketch Tool button.

6. On the Stereo View toolbar, click the Auto Toggle 3D Floating Cursor button, then move the cursor into the Stereo Window where it immediately transitions to the 3D Floating Cursor.

When you are in Auto Toggle 3D Floating Cursor Mode, it isn’t necessary to toggle in and out of the Stereo Window (using the F3 key). An important distinction must be made here. It should be understood that, once in the Stereo Window, the mouse can no longer be thought of as a typical 2D mouse. It now controls the 3D Floating Cursor in three dimensions. It is also worth noting that, when in Auto Toggle 3D Floating Cursor Mode, the images are fixed. That is, you adjust the elevation value using the separation of the 3D Floating Cursor rather than the parallax of the images.

the 3D Floating Cursor appears to rest oportion of the building in the 2-Pane Viewhich is approximately 456 meters). If you are new to working in stereo, youuse the 2-Pane View frequently. It showsleft and right images of the image pair. Wposition of the 3D Floating Cursor in thematches the identical position in the righat the correct X, Y, Z location.

43

5

6

7

51

Col lect ing the polygon feature 3. Examine the ArcMap display—you should now see a llected.

. Any re immediately

nually Toggle

ture

the same area,


1. Using the left mouse button, to click to collect the building vertices (the HOUSE layer is a polygon feature layer) around the roof’s perimeter. Always remember that you must make accurate collections in X, Y, and Z. A good strategy is to move close to the building corners, and then adjust Z with the mouse scroll wheel to get the correct X, Y, Z location.

2. Double-click to finish collecting the polygon (or press F2 on the keyboard).

2D display of the 3D feature you just co

The ArcMap display updates in real timecollections made in the Stereo Window areflected in the ArcMap display.

4. On the Stereo View toolbar, click the Ma3D Floating Cursor button.

Collecting another polygon fea

The next polygon you’ll collect is located inbut has a different elevation.

2

3

4

LYST FOR ARCGIS

Locat ing the polygon feature Col lect ing the polygon feature

and down until t image of the ppears to rest ane View. The

ters. the corners of

ding (or press

ures, do so.

USING STEREO ANA52

1. In the X window of the 3D Position Tool dialog, type the X coordinate 691119.9.

2. Type the Y coordinate 191376.2.You don’t need to enter a value in the Z window.

3. Click Apply on the 3D Position Tool dialog.

Prepar ing to col lect the feature

1. On the Stereo View toolbar, click the Fixed Cursor Mode button.

2. Notice that clicking the Fixed Cursor Mode button disables the Auto Toggle 3D Floating Cursor Mode. In Fixed Cursor Mode, the 3D Floating Cursor does not appear to be moving as it was in Auto Toggle 3D Floating Cursor Mode. Instead, the elevation of the 3D Floating Cursor is controlled by moving the images, which is also known as adjusting parallax.

3. Move the cursor over the Stereo Window and click inside the window. Then press F3 on the keyboard to activate the 3D Floating Cursor in the Stereo Window.

1. Move the scroll wheel on the mouse up the roof overlaps in both the left and righimage pair, and the 3D Floating Cursor aon the same portion of the roof in the 2-Pbuilding’s roof is approximately 457 me

2. Click to select vertices corresponding tothe building.

3. Double-click to finish collecting the builF2 on the keyboard).

4. If you want to collect other building feat

2

3

1

12

3

53

5. Press the F3 key to toggle off the 3D Floating Cursor. 6. Click Open on the Raster or TIN Surface dialog.tion option lgorithm to ure of interest. sing image

following is le to manually

ursor tab.

dialog to accept

8


Collecting a polyl ine feature

Like polygon features you collect using Stereo Analyst for ArcGIS, polyline features have an elevation component. In the next example, you’ll see how you can use the Terrain Following Mode to simplify collection of a road feature.Sett ing Terrain Fol lowing Mode opt ions

In this portion of the exercise, you’ll use Snap To Ground and the Terrain Following Mode in conjunction with a raster elevation data source.1. On the Stereo Analyst toolbar, click the Stereo Analyst

dropdown list and choose Options.

2. Click the Terrain Following Cursor tab of the Stereo Analyst Options dialog.

3. In the Elevation type section, check the option to Use external elevation information.

4. Select the Raster or TIN surface option, then click the Open button.

5. Navigate to \ArcTutor\StereoAnalyst\DEM and select the Altdorf_1m_dem.img raster surface.

You could specify the Use image correlainstead, which would use a correlation aplace the 3D Floating Cursor on the featFor more information about that, see “Ucorrelation” on page 138.

7. Make sure that Apply continuous terrainnot enabled since you still want to be abnavigate in Z.

8. Click Apply on the Terrain Following C

9. Click OK on the Stereo Analyst Options the changes and close the dialog.

1

2 3 4

97

LYST FOR ARCGIS

Locat ing the poly l ine feature 4. Notice the current elevation of the 3D Floating Cursor, o Window.

p the 3D

ion of the 3D ation ion is obtained fied on the eo Analyst

USING STEREO ANA54




Col lect ing a poly l ine feature

In this part of the exercise, you’ll be collecting a 3D line feature for the PAVED_ROAD layer. You should still be in Fixed Cursor Mode, and the Manually Toggle 3D Floating Cursor Mode should still be active. 1. Click the Target dropdown list and choose

PAVED_ROAD.

2. Move your cursor into the Stereo Window and click, then press the F3 key on the keyboard to activate the 3D Floating Cursor.

3. Move in X and Y to the part of the road just within the image pair overlap boundary.

which displays at the bottom of the Stere

5. Press the “s” key on the keyboard to snaFloating Cursor to the ground.

Using Snap To Ground moves the elevatFloating Cursor to the ground level elev(approximately 454 meters). This elevatfrom the elevation data source you speciTerrain Following Cursor tab of the SterOptions dialog.

2

3

1

1

4 3

55

6. Notice that the elevation of the area beneath the 3D 10. Continue to collect the road feature until you reach the d Y: 191446.9. (or press F2 on

errain

ng “c” on the

D Floating

9


Floating Cursor updates in the lower portion of the Stereo Window.

7. Press the “t” key on the keyboard to enter Terrain Following Mode.

Entering Terrain Following Mode forces the 3D Floating Cursor to constantly acquire an elevation value from the elevation source. Using the Terrain Following Mode is the same as constantly applying the Snap To Ground operation. Notice that, on the Stereo View toolbar, the Terrain Following Mode button is toggled on.

8. While still in Terrain Following Mode, begin collecting vertices for the road feature.

9. As you digitize, notice the CE90 and LE90 readings and associated colorblock at the bottom of the Stereo Window.

A green colorblock indicates that the 3D Floating Cursor is on the ground or feature of interest; a red colorblock indicates that the 3D Floating Cursor is not resting on the feature. If the colorblock is red, adjust the position of the 3D Floating Cursor slightly until it changes to green, then collect the next vertex.For more information about LE90 and CE90, see “Checking accuracy of 3D information” on page 146.

approximate coordinates X: 691225.9 an11. Double-click to finish collecting the road

the keyboard).12. Press “t” on the keyboard to toggle off T

Following Mode. 13. Toggle off Fixed Cursor Mode by pressi

keyboard.14. Press F3 to toggle off Manually Toggle 3

Cursor Mode.

6

7

10

LYST FOR ARCGIS

If you want to see the new feature in the ArcMap data 2. On the Editor toolbar, click the Target dropdown list and

ting Cursor

ss the F3 ually Toggle

d in the und [“s”] and individual spot

the area.

2

USING STEREO ANA56

view, you can click the Synchronize Geographic Displays button on the Stereo View toolbar.

Collecting point features

You collect point features, each of which has an X, Y, and Z coordinate, with a single click. You can use keyboard shortcuts and other tools you’ve learned about so far in the collection of point features.Locat ing the area for point features




4. Click Close on the 3D Position Tool dialog.Prepar ing to col lect point features

Spot heights are point features; therefore, one click both starts and completes digitizing a spot height.1. Use a combination of the Fixed Cursor Mode and the Z

thumb wheel to adjust the overlap of the images so that features overlap. Make sure you exit Fixed Cursor Mode when you are finished.

choose SPOTHEIGHT.

3. Make sure the Manually Toggle 3D Floabutton is selected.

Col lect ing point features

1. Click inside the Stereo Window, then prebutton on the keyboard to toggle on Man3D Floating Cursor Mode.

2. Use a combination of procedures outlineprevious examples (such as Snap To GroTerrain Following Mode [“t”]) to collectheights with a single click.

3. Collect approximately 10 spot heights in

2

3

1

4

4

57

r edits.

: Editing d the feature Analyst for

o closes Stereo

sting features”, nd move entire


4. If you used it, make sure that you have exited Terrain Following Mode. (The Terrain Following Mode button does not appear recessed on the Stereo View toolbar.)

5. Press F3 to toggle off Manually Toggle 3D Floating Cursor Mode.If you want to see the new features in the ArcMap data view, you can click the Synchronize Geographic Displays button on the Stereo View toolbar.

Saving features

1. Click the Editor dropdown list and choose Stop Editing.

2. Click Yes on the Save dialog to save you


If you plan to proceed directly to “Exercise 5existing features”, keep the raster images andatasets displayed in the ArcMap and StereoArcGIS applications.Otherwise, you may exit ArcMap, which alsAnalyst for ArcGIS.

What’s next?

In the final exercise, “Exercise 5: Editing exiyou’ll learn how to edit individual vertices afeatures in the Stereo Window.

2

1

2

LYST FOR ARCGIS

Exercise 5: Edi t ing exist ing featuresolder named

ip2_1.img and

o Window and

ting you about ion.n ArcMap es” on page 20 ges’ display.

ata button.

lder..mdb.

s.ers HOUSE, _TANKS.

E feature .img, so make ool to find the

ale dropdown

USING STEREO ANA58

Feature editing in Stereo Analyst for ArcGIS differs from traditional ArcMap feature editing by operating in 3D. In Stereo Analyst for ArcGIS, existing features can be edited using a 3D digital representation of the earth’s surface (created using overlapping, oriented images) as a reference backdrop for updating the existing dataset.While it is still possible to edit features solely in X and Y, you can no longer ignore elevation. All data collected and edited in Stereo Analyst for ArcGIS is 3D. In Stereo Analyst for ArcGIS, each vertex of every feature has an X, Y, and Z (elevation) value associated with it.This section focuses on editing polygon features, which involves editing vertices associated with a polygon and moving an entire polygon. The same editing procedures in Stereo Analyst for ArcGIS can be used regardless of the type of feature data (point, line, or polygon).

Preparing

This exercise assumes you’re using a standard computer mouse (with a scroll wheel).If you’re continuing the tutorial from “Exercise 4: Collecting features in 3D”, then you can proceed to “Adjusting a polygon feature” on page 58.If you’re starting this exercise from scratch, you should have an empty ArcMap data view and Stereo Window displayed. Also, you should have the Stereo Analyst, Stereo View, and Editor toolbars displayed. Then, proceed to “Adding images” below.

Adding images

1. Click the Add Data button to select the rasters.

2. In the Add Data dialog, navigate to the f\ArcTutor\StereoAnalyst\Images.

3. Ctrl-click to select the images named strstrip2_2.img.

4. Click Add to load the rasters in the StereArcMap.

5. If necessary, click OK on the dialog alerthe absence of spatial reference informatIf the display of the background values ibothers you, refer to “Changing propertifor instructions about how to fix the ima

Adding feature data

1. On the ArcMap toolbar, click the Add D2. In the Add Data dialog, navigate to the

\ArcTutor\StereoAnalyst\Geodatabase fo3. Double-click the file sampleAltdorfFME4. Click to select the layer named Building

The Buildings feature dataset has the layHOUSE_EXTENSION, and STORAGE

5. Click Add on the Add Data dialog.

Adjusting a polygon feature

Locat ing an exist ing polygon feature

The first feature to be edited is in the HOUScategory in image pair strip2_1.img/strip2_2sure it’s active. You’ll use the 3D Position Tfirst feature quickly.1. On the Stereo View toolbar, click the Sc

list and select 150%.

59

nchronize iew in the the Stereo


2. On the Stereo View toolbar, click the 3D Position Tool button.




After the new coordinates are applied, the portion of the image pair shown in the Stereo Window changes. Also, the 3D Floating Cursor moves to the coordinate location you entered. As you can see, two vertices (which are circled in yellow in the following picture) for this feature are not positioned on the building corners.

6. On the Stereo View toolbar, click the SyGeographic Displays button so that the vArcMap data view approximates that in Window.

1

2

3

4

56

LYST FOR ARCGIS

ss F3 on the ursor. ow, this cursor

s a regular 2D

er Fixed Cursor

just parallax so e elevation as 63 meters).

de after you

USING STEREO ANA60

Prepar ing to adjust the polygon feature

1. Click the Editor dropdown list and choose Start Editing.

2. On the Editor toolbar, set the Task to Modify Feature.3. On the Editor toolbar, set the Target to the HOUSE

layer.

4. Click the Edit Tool button.

5. Click the Manually Toggle 3D Floating Cursor button.

6. Click inside the Stereo Window, then prekeyboard to toggle on the 3D Floating CRemember that, while in the Stereo Windis a 3D cursor that no longer functions aWindows cursor.

7. Press the “c” key on the keyboard to entMode.

8. Using the scroll wheel on the mouse, adthat the 3D Floating Cursor is at the samthe roof of the building (approximately 4

9. Press “c” again to exit Fixed Cursor Mohave removed parallax.

Orient ing the ArcMap display

To orient the ArcMap display to match the display in the Stereo Window, select the Stereo Analyst dropdown menu on the Stereo Analyst toolbar, then click Options. On the ArcMap Display tab, click the check box for Orient ArcMap document to Image Pair when Image Pair changes. Then click OK.

1

1 3

4

5

61

right-most

ation of the 3D t erroneous

ges to a squared tion, when you


If you look at the 2-Pane View at the bottom of the Stereo Window, you’ll know you’re at the correct elevation when the 3D Floating Cursor is in the same position in both panes.

10. When the 3D Floating Cursor is at the correct elevation, position it in the middle of the polygon and click the left mouse button once. This selects the polygon. Notice how a dark blue square is positioned at the center of the polygon in the Stereo Window, and the polygon outline color is light blue in ArcMap.

Adjust ing vertex posi t ion

1. Position the 3D Floating Cursor over theerroneous vertex.

2. Using the scroll wheel, adjust the Z elevFloating Cursor to match that of the righvertex (approximately 467 meters).

3. Notice how the 3D Floating Cursor chancrosshair, shown in the following illustraare within the snap radius of a vertex.

8 10

LYST FOR ARCGIS

The following series of steps makes use of the 2-Pane

rrect X, Y, and Notice that the ts the old new location. the location of at an incorrect

tices locations, es a dialog with

that the are in the

USING STEREO ANA62

View to ensure that the 3D Floating Cursor is located at the same position in both the left and right images.

4. To move the vertex, position the 3D Floating Cursor within the snap radius of a vertex, then click and hold the left mouse button.

5. Move the 3D Floating Cursor in X and Y while holding down the left mouse button until its location corresponds to the appropriate corner of the building.

6. While continuing to hold the left mouse button, use the scroll wheel on your mouse to adjust the elevation of the 3D Floating Cursor. The elevation of the building is approximately 463 meters.

7. When the 3D Floating Cursor is in the coZ location, release the left mouse button.original feature outline (blue) still refleclocation, even though the vertex is in its

8. Using the same steps, proceed to correctthe left-most erroneous vertex, which is elevation of approximately 470 meters.

9. If you are satisfied with the adjusted verclick the right mouse button. This launchseveral options.

10. Choose the Finish Sketch option. Noticepolygon outlines update and the verticescorrect locations.

4

5

7

10

63

11. Click outside the polygon to exit feature editing. You don’t need to enter a value in the Z window. log.

og.ncorrectly

1

2

4


12. Press F3 to toggle off Manually Toggle 3D Floating Cursor Mode.When you’re done, the building should look similar to the one shown in the following picture.

Moving a polygon feature

In this section of an exercise, you’ll learn how to move a polygon in X, Y, and Z.Locat ing an exist ing polygon feature


2. Type the Y coordinate 191202.5.

3. Click Apply on the 3D Position Tool dia

4. Click Close on the 3D Position Tool dialYou’ll notice a house polygon which is ilocated offset from the actual building.

3

LYST FOR ARCGIS

Prepar ing to move the polygon feature 4. Click and hold the left mouse button, and move the in the X and Y

cation by the 2-Pane

ing Cursor o need to make

USING STEREO ANA64

1. On the Editor toolbar, click the Task dropdown list and select Reshape Feature.

2. On the Editor toolbar, confirm that the Target layer is still HOUSE.

Adjust ing polygon posi t ion

1. Click in the Stereo Window, then press F3 to toggle on Manually Toggle 3D Floating Cursor Mode.

2. Adjust the position of the 3D Floating Cursor using the scroll wheel on your mouse so that it rests on the roof of the house polygon (approximately 461 meters). This is easiest to accomplish at the corner of the building. You could also use the “s” keyboard shortcut to snap the 3D Floating Cursor to the roof.

3. Left-click inside of the misplaced polygon to select it. The polygon outline color turns to light blue. This means that the polygon is selected.

polygon to the correct building location direction. Again, it is easiest to judge the correct lolooking at the corners of the building in View.Since you already adjusted the 3D Floatelevation to the building’s roof, there is nfurther adjustments in Z.

5. Release the left mouse button.

1 2

2

4

65

6. Once the polygon has been placed in its correct position,

Map document

reo Analyst for

1


left-click anywhere outside the polygon to exit the editing task.

7. Press F3 to toggle off Manually Toggle 3D Floating Cursor Mode.If you want to see the new feature position in the ArcMap data view, you can click the Synchronize Geographic Displays button on the Stereo View toolbar.

Saving feature modifications

1. Click the Editor dropdown list and choose Stop Editing.

2. Click Yes on the Save dialog to save your edits.


1. Click the File menu in the ArcMap window, then click Exit.

2. Click Yes to save your changes to an Arcif you want; otherwise, click No.

Both the ArcMap application and the SteArcGIS application exit.

Edit ing a number of features

The same procedure described here, in “Moving a polygon feature” on page 63, can be used to edit a number of polygons that are displaced by the same amount in the X, Y, and/or Z. direction.

1

2

2

LYST FOR ARCGIS

What’s next?

USING STEREO ANA66

This tutorial has introduced you to some of the basic functions you can perform using Stereo Analyst for ArcGIS. The following chapters go into more detail about each element of the Stereo Analyst for ArcGIS suite of tools, and include instructions on how to use them to your advantage.

Wor

ion 2

king in stereo

Sect

LYST FOR ARCGIS
USING STEREO ANA68

69

3Wor

rate and reliable nted images: em into the

IN THIS C

• Creating

• Using IMcreate o

• Using Imto create

• Importinblock fil

• Importin

3
king with oriented images
Oriented images serve the most important role in collecting accuinformation from imagery. In this chapter, you’ll learn about oriewhere they come from, how to create them, and how to import thArcGIS environment.

HAPTER

oriented images

AGINE OrthoBASE to riented images

age Analysis for ArcGIS oriented images

g IMAGINE OrthoBASE es

g SOCET SET® files

LYST FOR ARCGIS

Creat ing or iented images

ed images.

us illustration, with

70

To undused to

Star t

On a daplaces,subseqwe mai

Subconfeaturerelationcan be

Now, a

Two rasensor,image stime th

• Fe• Re

off• Pro• Inf

USING STEREO ANA

lationships between features such as distance from the post ice to your officecesses such as the amount of water flowing in a riverormation such as the monetary value of a foreststand

This is the same relative area as depicted in the previoonly feature outlines displayed.

erstand an oriented image, it is helpful to look at the process create one.

ing with raw imagery

y-to-day, minute-to-minute basis, our eyes record people, and interactions. The images our eyes record are uently associated with features, relationships, and processes ntain in our brains.

sciously, we automatically associate intelligence with a . Once intelligence has been associated with a feature, ships and processes between that feature and other features deduced.

pply this concept to imagery.

w and overlapping images can be derived from an airborne satellite sensor, or even a hand-held digital camera. A raw erves as a permanent record of the state of the world at the

e image was captured. Images record:

atures such as houses, roads, rivers, schools

A Stereo Window can display features in a set of orient

WOR 71

Rawof thpermstore

The to

ographical ale, or how it was simple. ready since it

image, you have intelligent image s defining the n the image was age has

arth’s surface can they can be stored ships between een features are heir relationships

rom imagery, but e oriented image. an image must be e is the map of the

ssociation

fining the 3D the earth’s gulation (AT) and ducts such as

ese products are ep processes that n intelligent,

KING WITH ORIENTED IMAGES

p image is a simple TIF image; the bottom is a stereo view of the area.

oriented and made intelligent. An intelligent imagfuture.

Understanding image-to-ear th a

An image-to-earth association is the result of demathematical relationship between an image andsurface. This process is referred to as aerial trianis commonly done in digital photogrammetry proSOCET SET® and IMAGINE OrthoBASE®. Thfactories since they provide a series of step-by-stare required to transform the stupid image into aoriented image.

imagery is an untapped resource for creating and updating all e data used in a GIS. Stereo Analyst for ArcGIS transforms the anent record stored in a raw image into information directly d in the geodatabase.

A raw image knows nothing about geography, gerelationships, processes on the earth’s surface, screcorded—it’s a raster with pixels in it, plain andEssentially, it’s a “stupid” image that is not GIS-knows nothing about geography.

To get accurate, valuable information from a rawto process it to make it “intelligent”. Creating aninvolves associating it with the earth. This meanrelationship between an image (as it existed wherecorded) and the earth’s surface. Once a raw imintelligence associated with it, it is GIS-ready.

By making an image intelligent, features on the ebe collected. When features have been collected, within a database in multiple layers, and relationfeatures can be defined. Once relationships betwdefined, processes associated with features and tto other features can be derived.

Multiple levels of information can be extracted fthe foundation for information extraction is in thPrior to extracting any information from imagery,

LYST FOR ARCGIS72

The f

Onceinforavail3D mimagvaria

Pro

Insidis buintegimagto be

An ointell

sensor model.

tionship, is

telligence

ensor Model

ation

USING STEREO ANA

bles characterized by a sensor model.

cessing inside the factory

e the factory, either IMAGINE OrthoBASE or SOCET SET® sily adding intelligence to the raw image. This involves rating information from a variety of sources to compute the e-to-earth association. The following deductions can be made tter understand an oriented image.

riented image is defined as a stupid raster image plus igence.

A sensor model, which is a 3D mathematical reladefined by an image-to-earth association.

Sensor Model =

Image-to-earth Associ

actory adds information to the stupid image to make it intelligent.

a raw image has been transformed inside the factory, mation such as projection, units, and a sensor model is able for each intelligent, oriented image. A sensor model is the athematical relationship between the sensor used to record the e, the ground, and the image itself. These are the three general

Modified from Asher and Adams, 1976

Intelligence is defined as spatial reference plus a

Oriented Image =

Stupid raster image

+ In

Intelligence = +

Spatial Reference(projection/units)

S

WOR 73

The isatelimag

The iconsrelatigrou(X, Yusedare a

round control establish the istics associated t be defined. This s defining sensor

Imaging and tereo Analyst for ages provided by

to as rational ients that contain

e image and the ta are most

triangulation data SE. SOCET ll of the metadata rthoBASE block

create oriented


mage-to-earth association yields a transformation that ists of a series of coefficients describing the 3D mathematical onship between the image, the sensor that captured it, and the nd it has recorded. This process estimates the exact 3D position , Z) and rotation (three rotation angles) of the sensor that was

to record the image at the time of capture. These parameters lso referred to as exterior orientation parameters.

SET creates project and support files that store arequired to create an oriented image. IMAGINE Ofiles serve as metadata storage containers used toimages.

mage-to-earth association is defined as the position of the lite at the time of image capture, plus the image, plus the e’s location on the earth.

Image-to-earthAssociation =

+

+

These variables are computed by measuring 3D gpoints (GCPs) within the raw image. In order to image-to-earth association, the internal characterwith the sensor as reflected on the raw image musis referred to as interior orientation. This involveproperties such as focal length.

Defining an oriented image

Satellite images recorded and prepared by SpaceDigitalGlobe are referred to as oriented images. SArcGIS can directly use overlapping oriented imSpace Imaging and DigitalGlobe.

These oriented images contain metadata referredpolynomial coefficients (RPCs). RPCs are coefficinformation defining the relationship between thearth’s surface. Both the imagery and the metadacommonly stored in NITF or GeoTIFF format.

Other examples of oriented images include aerial created by SOCET SET® or IMAGINE OrthoBA

®

LYST FOR ARCGIS

s all of the

ccuracy of feature y used to assess

74

Unify

A GIS relation

By tracdata dewhat im

USING STEREO ANA

ing images, features, relationships, processes, and information

serves as a container for the feature datasets that have been extracted from oriented images. A GIS also maintainships, processes, and information associated with a feature dataset.

ing the ancestry of spatial information, it is evident that the reliability of information in a GIS is dependent on the arived from oriented imagery. The following example illustrates the ancestry of information derived from imagerpact a residential housing development may have on a watershed drainage system.

WOR 75

In tha GISthe oasses(localayeris, usrelatithe g

ArcGinforhydrand qrequiaccudoneaccu


is example, a factory is used to transform a raw image to create -ready, oriented image. Stereo Analyst for ArcGIS then uses

riented imagery to extract all of the feature data required to s the impact on the drainage system. This includes vegetation tion and type), roads, rivers, trees, and terrain. An additional of information can be derived from each feature dataset (that e Spatial Analyst™ to delineate a drainage system), and onships can be built between multiple layers of information in eodatabase.

IS uses these feature datasets in combination with other mation layers (such as annual precipitation) to conduct ological analysis to assess the overall impact on water quality uantity on the watershed of interest. Without imagery, the data red to assess the impact would not be available. Without rate and up-to-date feature data, a reliable study could not be . This example illustrates how intelligent information relies on rate, oriented imagery.

LYST FOR ARCGIS

Using IMAGINE OrthoBASE to create or iented imagesolves associating e. In IMAGINE

alibration. The AGINE l information and

BASE, first Ortho dialog opens.

ASE.

ge at a time or . In order to create he OK to all rlapping oriented ollection and

76

IMAGIfactorygenerat

In ordemust coimager

This is t

The pro

1. A2. D3. M4. Pe5. Pe

Once thinformIMAGIa time,

USING STEREO ANA

efining the sensor and properties associated with the sensor.easuring GCPs.rforming automatic tie point collection.rforming image-to-earth association (aerial triangulation).

e image-to-earth association has been completed, the ation required to create an oriented image is available. NE OrthoBASE allows you to create one oriented image at

or multiple oriented images simultaneously.

This is the Ortho Calibration dialog in IMAGINE OrthoB

You have the option of creating one oriented imacreating multiple oriented images simultaneouslymultiple oriented images simultaneously, select toption. It is important to note that at least two oveimages are required in order to perform feature cediting using Stereo Analyst for ArcGIS.

NE OrthoBASE can be thought of as a process-driven that creates oriented images and other first and second ion data layers. These layers are stored and used in a GIS.

r to create oriented images in IMAGINE OrthoBASE, you mplete the first five steps associated with processing raw

y.

he IMAGINE OrthoBASE interface.

cessing steps include:

dding images to your project.

The final process of creating an oriented image invthe sensor model metadata with the original imagOrthoBASE, this process is referred to as orthocoriginal image is not resampled or modified. IMOrthoBASE simply saves intelligent sensor modeassociates it with the original raw image.

To create the oriented image in IMAGINE Orthochoose the Process menu. From there, you selectRectification, then Calibration, and the following

WOR 77

In IMsourcorderthe Ielevaserverequitopognow ortho

Oncethe imArcMimag

For mrefer

Us

Stecamimacan

raster file from , the converted

st for ArcGIS, aster file, then oBASE block

aster file retains st for ArcGIS.


era, IKONOS, IRS-1C, QuickBird, and SPOT oriented ges created by IMAGINE OrthoBASE. Oriented images not be created for terrestrial or close-range images.

AGINE OrthoBASE, you are asked to define an elevation e. However, an elevation source is not always required in to create an oriented image, as in the case of importing using

MAGINE OrthoBASE and SOCET SET® importers. An tion source is required in this case since this interface also s the purpose of creating orthocalibrated images. This process res an elevation source to account for the effects of raphic relief displacement on the image. Thus, this interface

serves two purposes: creating oriented images and creating calibrated images.

the oriented images have been created, you can directly add ages to ArcMap for use in Stereo Analyst for ArcGIS. ap automatically recognizes that the images are oriented

es.

ore information on using IMAGINE OrthoBASE, please to the IMAGINE OrthoBASE User’s Guide.

ing data f rom IMAGINE OrthoBASE

reo Analyst for ArcGIS supports digital camera, frame

Using spat ia l database engine f i les

If you use the SDE converter to create an SDEa raster file calibrated in IMAGINE OrthoBASEfile will not retain the map model.

In order to use SDE raster files in Stereo Analyfirst convert the original raster file to an SDE rattach the SDE raster file to an IMAGINE Orthfile, then calibrate the file. The resulting SDE rthe map model and can be used in Stereo Analy

LYST FOR ARCGIS

Using Image Analysis for ArcGIS to create or iented imagesnalysis for

propriate sensor

mera properties

taset and the reference GCPs to the same

agery, this for exterior rs, this involves mage forming

pping raster

78

Anotheallows functiotoolbar

The proArcGISimage. using IIKONO

CreatinassociaThe oriinformStereo

The Geooriented

USING STEREO ANA

Correction tool in Image Analysis for ArcGIS allows you to create images that can be used by Stereo Analyst for ArcGIS.

5. Repeat steps 1 through 4 for the next overladataset.

r extension to ArcGIS, Image Analysis for ArcGIS, also you to create oriented images. You can access this nality via the Data Preparation menu in the Image Analysis .

cess of creating an oriented image in Image Analysis for overlaps with the process of creating an orthorectified The following models can be used to create oriented images mage Analysis for ArcGIS: camera, SPOT, QuickBird, and S.

g an oriented image in Image Analysis for ArcGIS involves ting sensor model information (metadata) to a raw image. ginal raw image is not modified, but the sensor model ation is added as metadata which is required and used by Analyst for ArcGIS for accurate feature collection.

In order to create oriented images using Image AArcGIS, the following steps must be completed.

1. Add raster dataset(s) to ArcMap.2. In the Image Analysis toolbar, select the ap

model.3. Within the GeoCorrection properties:

a. If using aerial camera imagery, define caand measure fiducial marks.

b. Define the elevation source.

c. Measure links between the raw raster dareference dataset. This involves locatingin the reference dataset and linking themlocation on the raw image.

d. Define sensor properties. For camera iminvolves defining initial approximations orientation, if known. For satellite sensodefining parameters associated with the isatellite sensor geometry.

4. Solve and save the solution.

79

Import ing IMAGINE OrthoBASE block f i lestension, and may strip of images eral strips of sociated with the , camera/sensor

d control point heroid, and datum

uirements

k file to determine the images are in e oriented and

ed by the block directory as the

ou’ll be prompted

alog.

ust click the Open hen click OK. If ation, they are y have to rename ou’ll be prompted l.

WORKIN

One ofto creatCreatedPro™, tnecessa

This is t

The imreferenthe direintelligbeen couse by

A blockcomes followi

Defin

A blockdata as

G WITH ORIENTED IMAGES

ced in the block file, verifies that the images are located in ctory specified by the block file, and then associates the ent metadata to the original images. Once this process has mpleted, the oriented images can be added to ArcMap for

Stereo Analyst for ArcGIS.

file, altdorf.blk, is included with the example data that with Stereo Analyst for ArcGIS. You can find it in the ng directory: \ArcTutor\StereoAnalyst\BlockFile.

ing a block f i le

file is data in a container storing all of the photogrammetric sociated with a strip or block of imagery.

This is the IMAGINE OrthoBASE importer file locator di

The dialog functions like any other file chooser. Jbutton and browse to locate the file in question, tall the other files in the block are in the same locautomatically attached to the block, then you onlthe block file. If other files are still unavailable, yto select the appropriate location for them as wel

the data formats you can use in Stereo Analyst for ArcGIS e oriented images is the IMAGINE OrthoBASE block file. in IMAGINE OrthoBASE or IMAGINE OrthoBASE he block file is a data container that stores all of the ry metadata information required to create oriented images.

he Import IMAGINE OrthoBASE Block File dialog.

port process opens the block file, identifies the images

Block files, which are binary files, have the .blk excontain information associated with one image, athat overlap or are adjacent to one another, or sevimagery. In the block file is all the information asblock including imagery locations on your systeminformation, fiducial mark measurements, grouninformation, image measurements, projection, spinformation.

Understanding block import req

Stereo Analyst for ArcGIS initially reads the blocthe location of the related images for importing. Ifthe location specified in the block file, they can bimported immediately.

If the files are not located in the directory specififile, Stereo Analyst for ArcGIS looks in the sameblock file. If the images are not in that location, yto locate them with the following dialog.

LYST FOR ARCGIS80

Becawhicthe b

Use t

USING STEREO ANA

use selecting a new image location changes the block file, h contains the location of the files, you’ll be prompted to save lock file with a new name in the following dialog.

his dialog to create a new block with the correct file location.

81

Importing an IMAGINE OrthoBASE block f i le

2

3

WORKING WITH ORIENTED IMAGES

You can import and orient block files derived from frame camera images, digital camera images, and sensors into Stereo Analyst for ArcGIS using the IMAGINE OrthoBASE block file importer.

1. On the Stereo Analyst toolbar, click the Stereo Analyst dropdown list and choose Import IMAGINE OrthoBASE Block File.

2. On the Import IMAGINE OrthoBASE Block File dialog, click the Open button and navigate to the directory containing the block file you want to import.

The Select images to orient window displays each image in the block file. You must select the images you want to orient. If you do not select any images, no importing occurs.

3. Use the Shift and/or Ctrl keys on the keyboard to select the files you want to orient and import from the IMAGINE OrthoBASE block file. You don’t have to select all of the images in the block file.

4. If you would like to view the images immediately, make sure the Add imported oriented images to ArcMap document check box is checked.

5. Click OK to start the import process.

The imported, oriented images display in the ArcMap data view and the Stereo Window.

1

4 5

LYST FOR ARCGIS

Import ing SOCET SET® f i lesou cannot import ocation, indicated _FILE_NAME in

ed as follows:

st\ o

+000+000+002+000+000+002+000+000+000+000+000+000

+006-002+000+000 +000N

+000+000+005+005

82

SOCETused toOnce aor SOCimages

This is t

A SOCinformThis filInterchThe prowith a pis what

A SOCexamplfind it ithat if ythan thwill ha

USING STEREO ANA

ation associated with a photogrammetric mapping project. e is an American Standard Code for Information ange (ASCII) file that is used as part of the import process. ject file contains general mapping information associated roject such as projection, units used, and so on. The .prj file

you select for import.

ET SET® project file, altdorf.prj, is included with the e data that comes with Stereo Analyst for ArcGIS. You can n the directory: \ArcTutor\StereoAnalyst\SocetSet. Note ou load the project and support example data in a place other e default, C:\arcgis\ArcTutor\StereoAnalyst\SocetSet, you ve to manually edit the SOCET SET® project and support

VERTICAL_REFERENCE 0A_EARTH 6.37813700000000eE_EARTH 8.18191912720360eELLIPSOID_CENTER 0.00000000000000e

0.00000000000000e0.00000000000000e

PROJECTION_TYPE UTM_PROJECTIOZONE 11FALSE_NORTHING_POS 0.00000000000000eFALSE_NORTHING_NEG 0.00000000000000eFALSE_EASTING_POS 5.00000000000000eFALSE_EASTING_NEG 5.00000000000000eGRID_NAME UTM_11NIMAGE_LOCATION escon

SET® is a digital photogrammetry software product that is create oriented images, DTMs, and orthorectified images. erial triangulation has been completed (using either ORIMA ET SET®), the information required to create oriented for use in Stereo Analyst for ArcGIS is available.

he Import SOCET SET Project File import dialog.

ET SET® project file (.prj) contains general project

files to indicate the correct location. Otherwise, yand orient the example files. You can correct the lby DATA_PATH in the project file and IMAGEthe support files, using a text editor.

A SOCET SET® project file is typically structur

PROJECT_FILE fDATA_PATH D:\Data\StereoAnaly

Socetset\escon_demCOORD_SYS 6XY_UNITS 1Z_UNITS 1MINIMUM_X_OR_LAT 0.00000000000000eMINIMUM_Y_OR_LON 0.00000000000000eMINIMUM_Z 2.00000000000000eMAXIMUM_X_OR_LAT 0.00000000000000eMAXIMUM_Y_OR_LON 0.00000000000000eMAXIMUM_Z 4.00000000000000eGP_ORIGIN_Y 0.00000000000000eGP_ORIGIN_X 0.00000000000000eGP_ORIGIN_Z 0.00000000000000eGP_SCALE_Y 1.00000000000000eGP_SCALE_X 1.00000000000000eGP_SCALE_Z 1.00000000000000eELLIPSOID WGS_84

WOR 83

The pimag(.supcontaraw i

This i

The iexterthis i

Un

SOimpwhlice


derstanding import ing l imi tat ions

CET SET® files in the geographic projection cannot be orted by Stereo Analyst for ArcGIS unless the system on

ich you’re working also has SOCET SET® installed and nsed.

roject file also contains references to raw images. Each raw e used in SOCET SET® has a corresponding support file ) associated with it. A support file is an ASCII file that ins the detailed photogrammetric metadata associated with a mage.

s part of a support file.

mport process accesses all of the metadata (interior and ior orientation parameters) associated with the image and uses nformation to create an oriented image.

LYST FOR ARCGIS

Importing a SOCET SET® project f i le

2

3

USING STEREO ANA84

You can import and orient SOCET SET® projects into Stereo

Analyst for ArcGIS using the SOCET SET® importer.

1. On the Stereo Analyst toolbar, click the Stereo Analyst dropdown list and choose Import SOCET SET Project File.

2. On the Import SOCET SET Project File dialog, click the Open button and navigate to the directory containing the project file you want to orient and import.

The Select images to orient window displays each file in the project file. You must select the files you want to orient. If you do not select any images, no importing occurs.

3. Use the Shift and/or Ctrl keys on the keyboard to select the files you want to orient and import from the SOCET SET® project file.

4. If you would like to view the images immediately, make sure the Add imported oriented images to ArcMap document check box is checked.

5. Click OK to start the import process.

The imported, oriented images display in the ArcMap data view and the Stereo Window.

1

4 5

85

What’s next?

WORKIN

In the nmethod2D To 3D data

G WITH ORIENTED IMAGES

ext chapter, “Working with 3D data”, you’ll learn about s for working with 3D data. These methods include Virtual 3D, conversion of 2D feature datasets to 3D, and exporting sets to 2D datasets.

LYST FOR ARCGIS
USING STEREO ANA86

87

4Wor

epresentation of w and viewed der to update S, the feature ng two feature dataset

t 3D, it must be

r virtual or set to 3D while completed, the ture data. The ates an entirely

nstant elevation oordinate for a the Z coordinate n the dataset.

IN THIS C

• Comparmodels

• Using V

• Setting

• Using th

• Using adoptions

• Using th

4
king with 3D data
Using two overlapping, oriented images, an accurate 3D digital rthe earth’s surface is created when displayed in the Stereo Windowith stereo hardware. Many existing GIS databases are 2D. In orexisting GIS datasets (2D or 3D) using Stereo Analyst for ArcGIdataset must be superimposed on the 3D digital earth created usioverlapping oriented images. In order to superimpose an existingon the 3D digital earth, the feature dataset must be 3D. If it is notransferred to 3D.

This process of transforming a feature dataset to 3D can be eithephysical. The virtual option temporarily transforms a feature dataworking in the Stereo Window. When the data updates have beenoriginal feature dataset is restored, but with more accurate 2D feaphysical option transforms an existing 2D feature dataset and crenew 3D feature dataset that both supports and contains 3D data.

Regardless of which option is used, an elevation source (either a coor external elevation file) is referenced to obtain a Z (elevation) cparticular X, Y coordinate derived from the feature dataset. Once source has been provided, it is then associated with each vertex i

HAPTER

ing 3D features and 3D

irtual 2D To 3D

Virtual 2D To 3D options

e 2D to 3D converter

vanced conversion

e 3D to 2D converter

LYST FOR ARCGIS

Comparing 3D features and 3D models

, and Z, but it also xample of a scene range, and

ciated with the ave a height 3D model in 3D

D feature datasets wever, Stereo

ion of 3D models.

88

Char

A 3D ffeatureof that vertex.house m191111

A 3D fea

A 3D febut it dmodel.

USING STEREO ANA

ture has X, Y, and Z coordinates for each vertex.

ature has 3D coordinate values associated with each vertex, oes not have volumetric information as is the case with a 3D

A 3D model has volumetric information.

A 3D model normally has a height attribute assofeature. For example, a 3D polygon feature can hassociated with it. In this case, it can be used as aGIS applications.

Stereo Analyst for ArcGIS allows you to collect 3such as 3D points, 3D lines, and 3D polygons; hoAnalyst for ArcGIS does not allow for the collect

acterizing 3D features

eature can be a 3D point, 3D line, or a 3D polygon. A 3D has an X, Y, and Z coordinate associated with each vertex feature. The Z coordinate is the elevation value of that For example, a vertex corresponding to the corner of a

ay have the X, Y, and Z coordinate values of 691402.4, .6, and 466.6, respectively.

Characterizing 3D models

A 3D model not only has 3D coordinates in X, Yhas volumetric information. The following is an ewith many 3D building models (shown in grey, oyellow).

89

Using Vir tual 2D To 3Ds that increase the e feature draping, to handle invalid g advanced

ArcMap, it won’t t is already in 3D. ate elevation on page 93 for

levation source, the feature layer corresponding include: constant ed irregular

ually, that is, your cGIS simply uses a constant he initialized Z

ata you display in Y information is

ure dataset to be Analyst toolbar nverter” on page

WORKIN

Virtualthe accactuallyimprov

Howevfeaturemust beon top order foinform

The Vito 3D ssurfacereferenlocatiovertex

The Vidatasetfor Z cwith eaprocessthe Ste2D (X afeature

To succto 3D, an elevelevatio

G WITH 3D DATA

only occurs when the feature dataset is being displayed in reo Window. Once all edits have been made and saved, only nd Y) coordinate information is written back to the original

dataset.

essfully perform the virtual conversion of a dataset from 2D you need (1) a list of feature classes for conversion and (2) ation source such as a constant elevation value or an external n file.

elevation information contained in a DTM file orelevation to project your feature datasets in 3D. Tvalue is for viewing purposes only. The feature dthe Stereo Window can be edited, but only X andsaved.

If you want the elevation information with a featretained, the Features to 3D option in the Stereo should be used. Refer to “Using the 2D to 3D co93 for more information about this capability.

2D To 3D conversion is useful when you want to improve uracy of an existing feature dataset, but are not interested in capturing 3D information from the dataset. That is, you can

e the X, Y coordinate values of feature vertices.

er, in order to improve the reliability and quality of the dataset, stereo feature collection and editing techniques used. As a result, the feature dataset must be superimposed

of the 3D digital representation of the earth’s surface. In r this to occur, the feature dataset must have 3D coordinate

ation associated with it.

rtual 2D To 3D capability temporarily transforms a dataset o that it can be superimposed on the 3D digital earth’s displayed in the Stereo Window. This is achieved by cing a user-defined elevation source at a particular X, Y n for Z coordinate information. The X, Y location of the is obtained from the original feature dataset.

rtual 2D To 3D function does not create a new feature . Rather, it simply references and queries an elevation source oordinate information and then associates that information ch vertex in the feature dataset. This Virtual 2D To 3D

If you need them, there are more advanced optionaccuracy of the conversion. These options includinclusion of planar features, and a choice of howelevations. Those options are discussed in “Usinconversion options” on page 95.

Note that if a 3D feature dataset has been added tobe considered for use in Virtual 2D To 3D since iIn this case, use the Features to 3D utility to updinformation. See “Using the 2D to 3D converter”additional details.

Understanding how it works

Once you define an input feature dataset and an ethe Z coordinate associated with a vertex node inis assigned the elevation value located within theelevation source. The supported elevation sourceselevation value, DEM, and ESRI-type triangulatnetwork (TIN) files.

The conversion of the data to 3D is performed virtdata is not actually edited. Stereo Analyst for Ar

LYST FOR ARCGIS

Sett ing Vir tual 2D To 3D opt ions

o 3D tab is where Stereo Analyst ertex of a feature

g file or a GRID model wherein alue associated

e that ng area. Using a elevation source on the earth’s

urce is selected, ic 2D geometry

ource.

2D feature forces ArcMap ) displayed in the the 3D Floating

2D feature layers ure datasets have es associated with

90

You ca2D To AnalysOptionTo 3D

The Virt

Usin

You cArcGIMAGfor A

USING STEREO ANA

ual 2D To 3D tab allows you to create temporary 3D feature data.

g TIN f i les

an only use ESRI-type TIN files with Stereo Analyst for IS. TINs generated in other applications, such as INE OrthoBASE Pro, cannot be used by Stereo Analyst

rcGIS.

datasets in the Stereo Window. This option also dynamic 2D geometry (such as the selection boxStereo Window to assume the same elevation as Cursor.

Selecting features

The Selected features list shows all of the currentin the ArcMap Table of contents. If multiple featbeen added to ArcMap, all of the 2D feature classthe feature datasets are shown in the list.

n define the options that control the application of Virtual 3D to your data. To access these options, click the Stereo t dropdown list on the Stereo Analyst toolbar, then click s. On the Stereo Analyst Options dialog, click the Virtual 2D tab.

Setting the elevation source

The Elevation Source section of the Virtual 2D Tyou define the reference elevation source used byfor ArcGIS to associate a Z coordinate with each vin a feature class.

A raster surface can be an ERDAS IMAGINE .imfile. In this case, the raster dataset is an elevationeach pixel in the raster dataset has an elevation vwith it.

A constant elevation value is a user-defined valuapproximates the elevation of the study or workiconstant elevation value is less accurate than an since it may not accurately reflect the topographysurface.

If either the Raster/TIN or Constant elevation sothen 2D feature datasets are displayed, and dynamis converted to 3D using the specified elevation s

Selecting the None option disables the display of

WOR 91

Set

The AOptioassocenabFor madva

KING WITH 3D DATA

t ing advanced options

dvanced Options button is used to access the Feature to 3D ns dialog. There, you can define detailed parameters iated with the 2D to 3D conversion. This button is only

led if a raster or TIN surface is selected as the elevation source. ore information on the advanced options, refer to “Using

nced conversion options” on page 95.

LYST FOR ARCGIS

Using Vir tual 2D To 3D

4

8

USING STEREO ANA92

The following workflow describes how to use Virtual 2D To 3D.

1. Start ArcMap and Stereo Analyst for ArcGIS.2. Add image pairs and 2D features to an empty data view.3. On the Stereo Analyst toolbar, click the Stereo Analyst

dropdown list and choose Options.4. On the Stereo Analyst Options dialog, click the Virtual 2D

To 3D tab.5. Click the option corresponding to the elevation source

you wish to use for virtual conversion.6. If you choose the Raster or TIN surface option, click the

Open button to select the file. If you select Constant elevation, type an elevation value.

Notice that the 2D features you added and displayed in step 2 are all listed in the Selected features window of the Virtual 2D To 3D tab.

7. If you’re using a Raster or TIN as the elevation source, click Advanced Options to make additional choices if you wish.

8. Click Apply on the Stereo Analyst Options dialog.9. Click OK on the Stereo Analyst Options dialog to close it.

3

5 67 9

Using advanced opt ions

The advanced options for Virtual 2D To 3D are explained in “Using advanced conversion options” on page 95.

93

Using the 2D to 3D converter

the feature data ed that source, s window. To lick method to ectively.

to give your 2D SRI TIN surface,

the input dataset iven the t file.

the conversion

WORKIN

The Coof inpu

• A a g

• A mu

• Stacla

• A

Geodata3D in St

Upda

While televatioexisting

The prothe inpuYou areto updafeature

G WITH 3D DATA

bases, feature datasets, and feature classes can be converted to ereo Analyst for ArcGIS.

t ing Z values of 3D feature datasets

he Convert Features to 3D dialog is most often used to add n information to 2D features, you can also use it to update 3D features with new elevation values.

cess to convert the data is the same—the only difference is t dataset. In this case, the data is already three-dimensional. supplying a raster or ESRI TIN file, or a constant elevation te the elevation (Z) component of each vertex of each .

The Convert Features to 3D dialog is where you begin process.

nvert Features to 3D option works with the following types t feature data:

workspace, which can consist of a collection of shapefiles or eodatabase;feature dataset within a geodatabase, which might contain ltiple feature classes;ndalone feature classes, which may consist of a feature ss stored within a geodatabase; andfeature class within a feature dataset.

Using the converter

Initially, you choose the 2D source that containsyou want to convert to 3D. Once you have selecteligible 2D features are listed in the Select classeselect them, you can use the Shift-click or Ctrl-cselect contiguous or noncontiguous classes, resp

In the Elevation Source section, you can choose data elevation from either an external raster or Eor a constant elevation that you supply.

The output dataset is placed in the same folder asunless you specify otherwise. The output file is gdesignation “_3D” to distinguish it from the inpu

LYST FOR ARCGIS


3

4

6

7

10

USING STEREO ANA94

The following is an example workflow of how to convert 2D features to 3D.

1. Start ArcMap and Stereo Analyst for ArcGIS.2. On the Stereo Analyst toolbar, click the Stereo Analyst

dropdown list and choose Features to 3D.3. On the Convert Features to 3D dialog, click the Open

button and browse to locate the 2D file.4. In the Select classes window, click to select the classes

you want to convert to 3D.5. Click the option corresponding to the elevation source

you wish to use for conversion.6. If you choose the Raster or TIN surface option, click the

Open button to select the file. If you select Constant elevation, type an elevation value.

7. Type a name for the Output dataset in the window if you don’t want the default name.

8. The check box next to Add converted 3D feature classes to ArcMap document is selected by default. Uncheck it if you don’t want the features to display after conversion.

9. Click Convert on the Convert Features to 3D dialog.10. Click Done when the status bar at the bottom of the

ArcMap window indicates the process is complete.

2

5

8

9

Naming the output dataset

When an input feature dataset is selected, the output dataset defaults to the original input dataset name with “_3D” appended to the name. You can change the default name by typing it in the Output dataset window. It defaults to the same directory and folder location as the input dataset.

95

Using advanced conversion opt ions

eature follows the in the height ting new vertices applying the ion. If a high- of the newly

ertices associated l feature dataset o the feature rce applied to the

sociated with the cing. In this e more vertices. or more include more

ct a raster or D conversion n’t have access

WORKIN

When ysettingswhich yconverprocessyou are3D and

The accaccuratparameOptioncan ma

The Fea

G WITH 3D DATA

ture to 3D Options dialog provides advanced settings.

original feature dataset.

When draping is turned on, additional vertices asdataset can be inserted as a function of point spainstance, the feature dataset is densified to includRefer to “Using thinning tolerance” on page 98 finformation about densifying a feature dataset tovertices.

ou convert features to 3D, you can accept the default that are provided on the Convert Features to 3D dialog, ou’ve already learned about in “Using the 2D to 3D

ter” on page 93, or you can use additional options in the ing of your data. These options are available to you whether using Virtual 2D To 3D or actually converting features to creating a new file.

uracy of the conversion process can be increased by using e elevation surfaces and by defining advanced conversion ters. These options are available to you on the Feature to 3D s dialog. This section describes each of those options so you ke the most appropriate selections for your data.

Using feature draping

By using draping, you set the condition that the fsubtle changes in elevation of the terrain surfacedimension. This is made possible by (1) interpolabased on the location of existing vertices and (2)elevation source used to perform the 3D conversresolution elevation source is used, the reliabilitysampled (interpolated) points is higher.

In the diagram on page 96, the points reflect the vwith the features. The bottom layer is the originaassuming a zero (sea level) elevation is applied tdataset. The top layer illustrates an elevation sou

Using advanced opt ions

Advanced options are only available if you seleTIN surface as an elevation source during the 3process. If you enter a constant elevation, you doto the advanced options.

LYST FOR ARCGIS
96 USING STEREO ANA
Draping off

Draping on

Z

Z

Original point

Interpolated point

97

oints is measured w is not exceeded

t

WORKIN

Usin

Point sin the swhen p

G WITH 3D DATA

g point spacing

pacing is the distance between points used (sampled) during the interpolation process. The distance between the pame units as the image pair displayed in the Stereo Window. The distance you specify in the Point spacing windooints are selected for interpolation.

Original point

Interpolated poin

80 Units 100 Units

80 Units 100 Units

Point spacing = 20

Point spacing = 10

Z

Z

LYST FOR ARCGIS98

Usi

The lfeaturedunthinntopog

By seto msecti

In thpointterm

Here,

Here,

associated with es are commonly

Options dialog, levation value is re class. For you may want to ll vertices tion value.

articular feature ion dropdown list n to select one of g the elevation um interpolated,

llowing sections. sociated with the cles representing assuming a zero set.

USING STEREO ANA

the point is outside the thinning tolerance and is retained.

the point is inside the thinning tolerance and will be eliminated.

sea level elevation) is the original feature dataset(sea level) elevation is applied to the feature data

ng thinning tolerance

ine thinning tolerance is only active when the Drape linear res on the terrain surface option is active. This option removes dant points contained within the feature dataset based on a ing tolerance defined by you. It’s useful when the variation in raphy is minimal.

tting a thinning tolerance, Stereo Analyst for ArcGIS checks ake sure that there aren’t any duplicate points in collinear ons. If you don’t want thinning, simply set the value to 0.

e following diagrams, the green circle represents the current , the black circles represent adjacent points, and the red line inating in an arrow represents the thinning tolerance.

Creating planar features

A planar feature is a feature in which all verticesthe feature have the same elevation. These featurflat features such as building roofs.

In the Planar Features section of the Feature to 3Dyou can select certain classes to which a single eapplied to all features contained within that featuexample, if a building feature is converted to 3D,constrain the building polygon to be flat so that aassociated with the polygon have the same eleva

The elevation value applied to each vertex for a pcan be determined in several ways. In the Elevatof the Planar Features section, you have the optiothe following techniques to be used for computinvalue: At centroid, Minimum interpolated, Maximand Average interpolated.

Each method of interpolation is described in the foIn each diagram, the points reflect the vertices asfeatures. The bottom layer (which has the red cir

WOR 99

Usin

The Aof thpixelshow

Usin

If yointervalueillust

Z

m

Z

ion

levations are e, then the largest e. The following ue.

n

levations are by the number of elevation is shows the

maximum value

maximum value

average value

minimum value

KING WITH 3D DATA

is used to assign the elevation value. The following ration shows the Minimum interpolated option.

inimum value

assigned to the feature. The following illustrationAverage interpolated value.

Z

g the At centroid opt ion

t centroid option takes the elevation from the physical center e feature, the centroid. For example, in a polygon, the center is used for the elevation value. The following illustration s the At centroid option.

g the Min imum interpolated opt ion

u select the Minimum interpolated option, elevations are polated for each vertex making up the feature, then the smallest

value at centroid

Using the Maximum interpolated opt

If you select the Maximum interpolated option, einterpolated for each vertex making up the featurvalue is used to assign the elevation to the featurillustration shows the Maximum interpolated val

Using the Average interpolated opt io

If you choose the Average interpolated option, einterpolated for each vertex, added, then dividedvertices, which yields an average elevation. This

Z

LYST FOR ARCGIS100

Res

In soconvprovinter

Usin

By dIn causes value

Usin

If yoeleva

Usin

Clickvalueyou elowe

USING STEREO ANA

nter 30, then invalid elevation values are assigned a value no r than 30 map units, such as meters.

olving invalid elevations

me instances, invalid elevations may exist during the 3D ersion and interpolation process. Stereo Analyst for ArcGIS ides several options to resolve invalid elevations during the polation and conversion process.

g the or ig inal e levat ion value

efault, the selected option is Keep the original elevation value. se of an invalid elevation value, Stereo Analyst for ArcGIS the original elevation value if the feature is 3D. The elevation s of any questionable points are not changed.

g a defaul t e levat ion value

u’re familiar with the terrain in your data, you can enter an tion value to apply to all questionable points.

g a minimum val id elevat ion value

the Use minimum elevation value check box and input the of the lowest possible elevation in your data. For example, if

101

Applying advanced 3D conversion

2

WORKING WITH 3D DATA

parameters

You begin the advanced 3D conversion process by selecting the 2D feature dataset and an associated elevation source.

1. On the Stereo Analyst toolbar, click the Stereo Analyst dropdown list and select Features to 3D.

2. On the Convert Features to 3D dialog, click the Open button and navigate to the directory that contains the 2D data.

3. On the Import Features dialog, select the data file and click Open.

1

LYST FOR ARCGIS

4. Position your cursor inside the Select classes window

4

5

6

8

USING STEREO ANA102

and choose the classes you want to convert to 3D. You may use the Shift and Ctrl keys on the keyboard to select neighboring and non-neighboring classes. Selected classes have a blue background.

5. In the Elevation Source section, either click the Open button and select a raster or a TIN file to use for elevation, or enter a constant elevation for conversion.

6. If you wish, click the Open button and navigate to a different directory for the output file (by default, the output file is placed in the same directory as the input file). Give it a different name if you wish.

7. By default, the Add converted 3D feature classes to ArcMap document check box is selected.

8. Click the Options button on the Convert Features to 3D dialog.

9. Click the check box for Feature Draping and type the Point Spacing and Thinning Tolerance values.

10. Click inside the Planar Features window and select the classes you want to use an interpolated value.

11. Specify how you want elevation to be assigned to questionable areas.

12. Enter a minimum valid elevation value if you wish.13. Click OK on the Feature to 3D Options dialog.14. Click Convert on the Convert Features to 3D dialog.15. Click Done when the status bar at the bottom of the

ArcMap window indicates that the process is complete.

7

9

10

11

14 15

12

13

103

Using the 3D to 2D converterool, refer to the page 31 in

WORKIN

Stereo your 3DConver3D conAnalyscompo

The Confrom fea

At the named if you wdifferenchange

G WITH 3D DATA

vert Features to 2D dialog allows you to remove the height attribute ture datasets.

end of the export process, a new file is generated that is like the input file, but with an “_2D” designation. Of course, ant that file to be named something else and placed in a t directory, you can click the Open button and make those

s in an Output Features dialog.

Analyst for ArcGIS also provides you with the ability to take feature datasets and easily convert them to 2D using the

t Features to 2D dialog. The process is the same as the 2D to verter—you simply select the input 3D dataset, and Stereo t for ArcGIS produces a 2D dataset without the elevation (Z) nent.

For complete instructions about how to use this tsection called “Converting 3D features to 2D” onchapter 2 “Quick-start tutorial”.

LYST FOR ARCGIS

What’s next?

104

Next, yStereo 1-Panelearn aband how

USING STEREO ANA

ou’ll learn how to view in stereo (3D). You do so in the Window, which has three possible configurations—the View, the 2-Pane View, and the 3-Pane View. You’ll also out the toolbars that come with Stereo Analyst for ArcGIS to best use the tools to your advantage.

105

5Visu

easily view and methods. This ir most effective

tereo viewing dow views.

tereo View, and

dless of whether

ay options for

IN THIS C

• Introduc

• Using S

• Learningtoolbar

• Learningtoolbar

• LearningEnhance

• Using Swith Arc

• Setting

5
alizing in stereo
Stereo Analyst for ArcGIS provides you with tools so that you caninterpret your data—you do this using typical stereo visualizationchapter introduces you to those methods and gives you tips for theuse in Stereo Analyst for ArcGIS.

In this chapter, you’ll learn about viewing imagery in 3D using stechniques. These techniques include using different Stereo Win

You’ll also learn how to use tools found on the Stereo Analyst, SStereo Enhancement toolbars.

In the Stereo Window, you’ll see how to apply various tools regaror not it is docked to or undocked from the ArcMap window.

Finally, you’ll learn about the best application of the stereo displyour application.

HAPTER

ing stereo visualization

tereo Window views

about the Stereo Analyst

about the Stereo View

about the Stereo ment toolbar

tereo Analyst for ArcGIS Map

stereo display options

LYST FOR ARCGIS

Introducing stereo v isual izat ion

ct have binocular n is referred to as

are obvious.

106

On a davision. stereos

This ana

USING STEREO ANA

ily basis, we unconsciously perceive and measure depth using our eyes. Persons using both eyes to view an objePersons using one eye to view an object have monocular vision. The perception of depth through binocular visiocopic viewing.

glyph image shows a 1:1 image pixel to screen pixel ratio. With red/blue glasses, the drastic elevation differences in the region

VISUA 107

Vie

Withwith brainphotousingsurfageogis beinsteimag

Steregreatothersuppdata

• SbwAsdur

• Sit

using stereo esn’t depend on a tions in depth ly transformed ed with raw are not

n compared to of an external (DSM) created ation model is not

tereo Analyst for tion of depth to curate 2D and 3D

viewing, see page 199.

LIZING IN STEREO

ensor model information) are required for the highly accurate etermination of 3D information. Sensor model information is sed together with imagery to create a 3D digital epresentation of the earth’s surface.ystematic errors associated with raw photography and

magery are considered and minimized during the block riangulation process.

wing imagery in 3D

stereoscopic viewing, depth information can be perceived great detail and accuracy. Stereo viewing allows the human to judge and perceive changes in depth and volume. In grammetry, stereoscopic depth perception plays a vital role in imagery to create and view 3D representations of the earth’s ce. As a result of viewing overlapping images in stereo, raphic information can be collected to a greater accuracy. This cause a true 3D representation of the earth’s surface is used ad of the traditional monoscopic techniques that use only one e.

o feature collection techniques using oriented imagery provide er GIS data collection and update accuracy when compared to data collection and capture techniques. The following reasons ort why stereo feature collection techniques provide greater capture reliability and quality:

ensor model information derived from photogrammetric lock triangulation processing eliminates errors associated ith the uncertainty of sensor model position and orientation. ccurate image position and orientation information (that is,

• The collection of 3D coordinate informationviewing and feature collection techniques doDEM as an input source. Changes and variaperception can be perceived and automaticalusing the sensor model information associatimagery. Therefore, DTMs containing errorsintroduced into the collected GIS data. Wheorthorectified images (which require the useelevation source), a 3D digital stereo model using imagery is more accurate since an elevrequired as input.

• Digital photogrammetric techniques used in SArcGIS extend the perception and interpretainclude the measurement and collection of acGIS feature data.

For more detailed information about stereoscopicappendix B, “Understanding stereo viewing” on

LYST FOR ARCGIS

Using Stereo Window views

ot support stereo y reverts to upport.erdas.com/orted for use with

in anaglyph mode.

cGIS displays the reo view.

108

Stereo configu

Usin

In this orienteorienterepreseapprop

Viewingemitter a

The 1-Pthe Ste

USING STEREO ANA

in quad-buffered stereo requires special hardware, such as an nd stereo glasses.

ane View button is located within the lower-left portion of reo Window.

If you do not have a stereo graphics card, you can work

In anaglyph, shown above, Stereo Analyst for Aroriented images in red, green/blue to create a ste

Analyst for ArcGIS provides three different Stereo Window rations for the collection of reliable GIS data using imagery.

g the 1-Pane View

view, the sensor model information associated with each d image in an image pair is used to visually superimpose the d images on one another, thereby creating a 3D digital ntation of the earth’s surface when viewed with the riate stereo viewing hardware.

If the graphics card used by the computer does nviewing, Stereo Analyst for ArcGIS automaticallanaglyph stereo mode. See the Web site <http://sspecs/specs.html> for a list of graphics cards suppStereo Analyst for ArcGIS.

VISUA 109

Usi

In thwith mono

The 2

The 2availyou cthis cboth and a

2-Pane View e Stereo Window.

ane View are guration was e Stereo Window right mono panes.

no images.

LIZING IN STEREO

-Pane View shows the left image and the right image of the image pair.

-Pane View is useful when stereo viewing hardware is not able for stereo visualization. The 2-Pane View is also useful if annot properly view overlapping oriented images in stereo. In ase, positioning the 3D Floating Cursor on the same feature in panes during collection allows you to collect GIS data reliably ccurately.

The 3-Pane View displays the stereo view and both mo

ng the 2-Pane View

e 2-Pane View, the left and right oriented images associated an image pair are separately displayed within the left and right panes of the Stereo Window, respectively.

The 2-Pane View can be enabled by selecting thebutton located within the lower-left portion of th

Using the 3-Pane View

In the 3-Pane View, the 1-Pane View and the 2-Pembedded within the Stereo Window. This confidesigned so that you can collect feature data in thwhile verifying data collected within the left and

LYST FOR ARCGIS110

If thebeingsameand rsimpand d

The 3Analthe 3Stere

Sw

If yousingView

Use oappethe le

USING STEREO ANA

f Invert Stereo Model is appropriate when areas of elevation ar recessed in the Stereo Window and vice versa. Switching ft and right images corrects this problem.

3D Floating Cursor has been accurately placed on the feature collected, then the 3D Floating Cursor should reference the geographic location for that feature collected within the left ight mono panes. The 3-Pane View is intended to provide a le setup for quick quality assurance during feature collection ata update.

-Pane View is the default Stereo Window setup used in Stereo yst for ArcGIS. The 3-Pane View can be enabled by selecting -Pane View button located within the lower-left portion of the o Window.

i tching left and right images

u need to switch the left and right images, you can do so by the Invert Stereo Model button, which is located on the Stereo toolbar.

111

Learning about the Stereo Analyst toolbar image pair to

VISUAL

The Stedisplay

.

IZING IN STEREO

reo Analyst toolbar is used to access functions and options associated with Stereo Analyst for ArcGIS, select the in the Stereo Window, and to open the Stereo Window.

Image Pairs listLets you choosewhich image pairto display in theStereo Window

Stereo WindowClick to open theStereo Window

Image Pair Selection ToolClick to select an image pairfrom the ArcMap data view

The Stereo AnalystDropdown MenuGives you access

to a number of functions

from a dropdown list

LYST FOR ARCGIS

Learning about the Stereo View toolbar

elageys as versa

res Toolsh theaturesWindow

112

The Ste

USING STEREO ANA

reo View toolbar contains a number of tools used to efficiently operate within the Stereo Window.

Zoom In ToolClick to magnifyby a power of 2

Zoom Out ToolClick to reduceby a power of 2

RoamClick to view

a different areaof the image pair

in real time

Scale ToolClick to select

a predefined scale

Fixed Cursor ModeClick so that thecursor remains

stationary

3D Position ToolClick to enter X, Y,and Z coordinatesof a particular point

Zoom to Data Extent

Default ZoomClick for a 1:1image pixel to

screen pixel ratio

Zoom In By 2Click for 1-time

application of zoom

Zoom Out By 2Click for 1-time

application of zoom

Click to display all data in the

Stereo Window

Auto Toggle 3D

Click to collectfeatures without toggling

Terrain Following ModeClick so that the 3D Floating

Cursor is automatically placedon the ground or feature

of interest

Invert Stereo ModClick so that the left im

of the image pair displathe right image and vice

Synchronize Geographic DisplaysClick so that the Stereo Window

and the ArcMap data view displaythe same data coverage

Manually Toggle3D Floating Cursor

Click to use theF3 key to turn

the 3D Floating Cursor on/off in the

Stereo Window

Refresh FeatuClick to refredisplay of fe

in the Stereo

Floating Cursor

and drive to thatlocation

113

Learning about the Stereo Enhancement toolbary displayed in the

VISUAL

The SteStereo

IZING IN STEREO

reo Enhancement toolbar provides tools that allow you to control the contrast and brightness levels of the imagerWindow.

Image SelectorClick to select

whether to applyenhancement tothe left, right, orto both images

of the image pair

Decrease BrightnessClick to lower theamount of light

in the image pair

Reset BrightnessClick to reset theamount of light

in the image pairto when it was

originally loaded

Increase BrightnessClick to raise theamount of light in the image pair

Thumb WheelIncrease brightness, right;decrease brightness, left

Thumb WheelIncrease contrast, right;decrease contrast, left

Decrease ContrastClick to lower the

apparent differencebetween light and dark

in the image pair

Increase ContrastClick to raise the

apparent differencebetween light and dark

in the image pair

Reset ContrastClick to reset the

amount of differencebetween light and dark

in the image pairto when it was

originally loaded

LYST FOR ARCGIS

Opening the Stereo Window

1

USING STEREO ANA114

The Stereo Window is where an image pair is displayed to create a 3D digital representation of the earth’s surface. In the Stereo Window, you can view your 3D features as well as update and digitize features accurately. To open the Stereo Window, perform the following steps.

1. Display the Stereo Analyst toolbar by clicking the ArcMap View menu, then Toolbars, then checking the Stereo Analyst toolbar.

2. Click the Stereo Window button on the Stereo Analyst toolbar to open the Stereo Window.

2Docking Stereo Analyst for ArcGIS windows

By default, when you first open a Stereo Window, it is incorporated into the ArcMap workspace. You can undock it, however, by clicking and holding on the bar at the top of the Stereo Window and then dragging it outside the ArcMap environment. The 2-Pane View works in the same fashion, but the bar that controls docking is on the left side.

Docking the Stereo Window outside of the ArcMap workspace is advantageous if a dual monitor configuration is being used for feature collection. In this scenario, the Stereo Window is used as the interface for collecting and editing features while the ArcMap workspace is used as the cartographic station for feature verification, attribution, and analysis.

115

Applying tools in the Stereo Window

VISUALIZING IN STEREO

Several tools designed to work in conjunction with the Stereo Window can be used to increase the performance and productivity of feature collection and editing.

1. Add data to the Stereo Window.2. On the Stereo Analyst toolbar, click the Stereo Window

button to open a Stereo Window.

The Stereo Window opens with the setup displaying the image pair in the top portion of the window, and the left and right images of the image pair in the bottom portion of the window—the 3-Pane View. In this example, the Stereo Window is undocked from the ArcMap application.

3. On the Stereo View toolbar, click the Roam Mode button.4. Move your cursor (which appears as a hand) into the

Stereo Window and hold down the left mouse button. While the cursor is in the Stereo Window, drag the image pair to a new position.

2

3

4

1

LYST FOR ARCGIS

5. Double-click inside the Stereo Window to enter the

USING STEREO ANA116

Continuous Roam Mode.

The cursor changes from a hand to an arrow. As you change the location of your mouse, the arrow changes direction and speed accordingly.

6. Double-click inside the Stereo Window to exit the Continuous Roam Mode.

You can also use the Zoom In/Out Tool in the Continuous Zoom Mode.

7. Click the Zoom Out (or Zoom In) Tool button on the Stereo View toolbar.

5

7

117

8. Click and hold the scroll wheel inside the Stereo Window,


then move your mouse towards and/or away from you to activate the Continuous Zoom Mode.

9. Release the scroll wheel to exit Continuous Zoom Mode.10. Click the Default Zoom button to return the display to a

1:1 default zoom.11. On the Stereo Enhancement toolbar, click the Adjust

dropdown arrow and select Right Image.12. On the Stereo Enhancement toolbar, click the Image

Brightness Wheel and drag it all the way to the right to change the display of the right image.

13. On the Stereo Enhancement toolbar, click the Image Contrast Wheel and drag it to the right to change the display of the right image.

8

10

11

12 13

LYST FOR ARCGIS

14. Observe the changes in the display.

USING STEREO ANA118

Usually, you won’t make such drastic changes to the appearance of the images in the Stereo Window. If you don’t like the results, you can always reset the display of the image pair to its original brightness and contrast.

15. On the Stereo Enhancement toolbar, click the reset buttons both for Brightness and for Contrast.

The display returns to the original settings.

15

119

Using Stereo Analyst for ArcGIS wi th ArcMap

an be set to the re collection u to synchronize hic Displays

aying the data at ocate the same ng is retained for pen in future e orientation

ir in the same the orientation k.

ument to Image play in the rotation. If you e Tools provided

ly selected and ted image

undary of an alyst for ArcGIS erlap, that region he overlap pair, but can also

VISUAL

Most oconjun

There aWindofeatureaccessiAnalysthe Arc

The Arcbehavio

IZING IN STEREO

Map Display tab is where you choose settings that control image pair r in the ArcMap data view.

Selecting image pairs

Within ArcMap, an image pair can be interactivethen displayed in the Stereo Window using orienfootprints.

A footprint is a graphical representation of the boindividual oriented image. By default, Stereo Andisplays the footprint in red. When two images ovis again represented graphically, but in yellow. Tportions of two oriented images is called an imagebe referred to as a stereo pair.

f the time, you’ll be using Stereo Analyst for ArcGIS in ction with ArcMap.

re a few advanced options you can set so that the Stereo w and ArcMap data view compliment one another during collection and editing. You can set these options by ng the Stereo Analyst toolbar, then clicking the Stereo t dropdown list. From the list, select Options, then click on Map Display tab.

Orienting displays

The Stereo Window and the ArcMap data view csame orientation to make visualization and featueasier. Stereo Analyst for ArcGIS also allows yothe two displays (using the Synchronize Geograpbutton on the Stereo View toolbar).

With both ArcMap and the Stereo Window displthe same resolution and rotation, you can easily lfeature in both applications. The orientation settiArcMap documents (.mxd) that you save and reoStereo Analyst for ArcGIS sessions. Similarly, thsetting is retained if you open different image paArcMap session. However, if you exit ArcMap, setting must be reset when you resume your wor

Note that if you uncheck the Orient ArcMap docPair when Image Pair changes check box, the disArcMap data view does not return to the originalwant the image pair to unrotate, use the Data Framby ArcMap.

LYST FOR ARCGIS120

If theusingStereArcMStereoverlsamecolor

Opt

In caimagoriendisplyou tthis operfo

Cal

SterepairsSincepositwhetimagidentfor st

Duripurppropreprebetwcommview

een two strips nt. An image pair strips is not ideal

0 and a minimum verlapping images. of 100 and a en choose the alue, there are more

USING STEREO ANA

each oriented image contains sensor model information (3D ion and orientation), Stereo Analyst for ArcGIS can determine her or not the two images overlap. In cases where multiple es are being used in a project, Stereo Analyst for ArcGIS may ify many image pairs, although not all of them may be suitable ereo viewing and feature collection.

ng the collection of raw aerial photographs, flight plans are osely coordinated so that resulting photographs along a strip erly overlap so as to create an optimum 3D digital sentation of the area of interest. The overlap percentage een two photographs in a strip (also referred to as endlap) is

only 60 percent. This type of image pair is ideal for stereo ing and feature collection.

The top, left illustration shows a maximum overlap of 10overlap of 60. With these settings, you can choose all oThe bottom, right illustration shows a maximum overlapminimum overlap of 25. With these settings, you can evsidelap area of image pairs. If you lower your minimum vpossible image pairs from which to choose.

se colors don’t suit your needs, you can easily change them the dropdown arrows on the ArcMap Display tab of the o Analyst Options dialog. The color changes are retained for ap documents (.mxd) that you save and reopen in future

o Analyst for ArcGIS sessions. Similarly, the footprint and ap colors are retained if you open a different image pair in the ArcMap session. However, if you exit ArcMap, the default s are reinstated when you resume your work.

imizing display performance

ses where a mapping project uses many images (more than five es) or large images (greater than 85 MB), the display of ted images in the ArcMap data view may be slow. Rather than ay each raster in ArcMap, Stereo Analyst for ArcGIS allows o display only the footprints of the oriented images. Selecting ption on the ArcMap Display tab improves the display rmance in ArcMap.

culating usable image pairs

o Analyst for ArcGIS automatically computes valid image by determining whether or not two oriented images overlap.

The overlap percentage of two photographs betw(also referred to as sidelap) is commonly 20 percecreated from two overlapping images in adjacentfor stereo viewing and feature collection.

VISUA 121

StereImagFor econsbe co

Depevalueworkyou wget im

LIZING IN STEREO

o Analyst for ArcGIS computes image pairs based on an e Pair threshold percent setting on the ArcMap Display tab. xample, if the threshold is set to 50 percent, the two images

idered must share at least half of the same geographic area to nsidered an image pair.

nding on the type of data you use, the minimum and maximum s may necessarily be different. For example, you might be ing with images that have only a 30 percent overlap; therefore, ould change the minimum threshold value to ensure that you age pairs you can use in Stereo Analyst for ArcGIS.

LYST FOR ARCGIS

Using the Stereo Window with ArcMap

USING STEREO ANA122

When you use the Stereo Window in conjunction with ArcMap, your ability to verify, assess, and analyze collected features improves. Two important tools are Orient ArcMap document to Image Pair when Image Pair Changes and Synchronize Graphic Displays.

1. You may already have an image pair displayed in ArcMap and the Stereo Window. This way, you can see the change in orientation more clearly.

2. Click the Zoom to Data Extent Button to see all of the image pair in the Stereo Window.

3. Note the position of the image pairs in ArcMap.

The overlap portion of the current image pair is represented by the yellow border; image footprints are represented by the red border.

If you are working in anaglyph mode, the overlap portion of the current image pair displays in gray color, and the left and right images of the image pair display in red and blue, respectively.

If your graphics card supports quad-buffered stereo, the entire display, both overlap and nonoverlap areas, appears gray, but you can only see in stereo in the overlap area.

You can get details about your graphics card by accessing the Stereo Analyst toolbar, then clicking the Stereo Analyst dropdown list, then clicking Graphics Card Information.

4. From the Stereo Analyst toolbar, click the Stereo Analyst dropdown list, and then select Options.

3

4

2

123

5. Click the ArcMap Display tab. 5 786


6. Check the Orient ArcMap document to Image Pair when Image Pair changes check box.

By choosing this option, you ensure that the rotation of the image pair in ArcMap matches that of the image pair displayed in the Stereo Window. As a result, the orientation of the image pair in the Stereo Window is used as a reference for orienting the ArcMap display. This may help you locate features in both the Stereo Window and the data view.

7. Click Apply on the Stereo Analyst Options dialog.8. Click OK on the Stereo Analyst Options dialog.9. Notice that the display of the image pair in ArcMap has

changed to reflect the same north-south orientation as that in the Stereo Window.

Now, when you use tools such as Synchronize Graphic Displays and Default Zoom, which are both located on the Stereo View toolbar, the Stereo Window and ArcMap both display the image pair in the same orientation. This can help you as you digitize features.

9

LYST FOR ARCGIS

Sett ing stereo display opt ions

ecked, removes tereo Window.

ed when the 3D nsures a more

e “Using Fixed

age in an image a 3D digital to optimally viewing A feature positioned along

along the epipolar feature appearing ear along the stereo and the eliable.

tereo Analyst for ade in order for

r both the left and

ng the epipolar

124

Like thArcMaorienteby acceAnalysthe Ste

This is t

Choo

By defapair in overlapselect tdisplay

USING STEREO ANA

he Stereo Display tab.

sing the display area

ult, Stereo Analyst for ArcGIS displays the entire image the Stereo Window. However, if you only want to see the region common to the two images of an image pair, you can he Image Pair overlap region option. This only changes the in the Stereo Window, not the ArcMap data view.

accuracy of the collected feature may not be as r

By selecting the Use epipolar correction option, SArcGIS determines what correction needs to be mthe feature to be positioned on the epipolar line foright images.

For more detailed information, see “Understandiline” on page 206.

e options you can set affecting the display of data in p, there are also options you can set that affect the display of d images in the Stereo Window. You can set these options ssing the Stereo Analyst toolbar, then clicking the Stereo t dropdown list. Select Options from the list, then click on reo Display tab on the Stereo Analyst Options dialog.

Applying epipolar correction

The Use epipolar correction check box, when chY-parallax from the image pair displayed in the SThis leaves only X-parallax, which can be adjustFloating Cursor is in Fixed Cursor Mode. This eaccurate and visually comfortable stereo view.

For more information on Fixed Cursor Mode, seCursor Mode” on page 156.

Using sensor model information, each oriented impair is displayed so that, when viewed in stereo, representation of the earth is perceived. In order perceive imagery and features in 3D using stereotechniques, the images must be properly aligned.appearing on both the left and right image must bethe epipolar line.

At times a feature may not be properly displayed line in both the left and right images. If the same in two overlapping oriented images does not appepipolar line, it is difficult to view the images in

VISUA 125

Mo

The sscreepitchand iscreeStere

To ddivideachof thStereInfor

InformInform

See tlist oArcG

, and actual an exact ble area on the

solution.

epresent raw data , the range of data ightness values of

the values of the esults in a more

ast stretches: Two

LIZING IN STEREO

ation about your system is readily available on the Graphics ation dialog.

he Web site <http://support.erdas.com/specs/specs.html> for a f graphics cards supported for use with Stereo Analyst for IS.

difying screen dot pitch

creen dot pitch is the size of the pixels on the screen. The X n dot pitch value is measured horizontally; the Y screen dot is measured vertically. This value is automatically computed s used for scaling in the Stereo Window. The more accurate the n dot pitch values are, the more accurate the scaling is in the o Window.

etermine the appropriate screen dot pitch size for your display, e the screen physical size by the screen pixel resolution for dimension. This information may be found on the Screen tab e Graphics Information dialog, which you access from the o Analyst dropdown menu by clicking Graphics Card mation.

The screen physical dimensions are approximatedimensions may vary due to monitor settings. If measurement is required, then measure the viewamonitor and divide by the current screen pixel re

Applying contrast stretch

Since the data file values in raster images often r(such as elevation or an amount of reflected light)file values is often not the same as the range of brthe display.

A contrast stretch, as the name implies, stretchesraster image to fit the range of the display. This rcrisp looking image in the Stereo Window.

You can choose from the following types of contrstandard deviations, Min/max, Linear, or None.

LYST FOR ARCGIS126

Usin

A Tw-2 anstretc

e data values vary maximum to the maximum

grey

USING STEREO ANA

Output dataHistogram is yellow

Input dataOriginal histogram is grey


Input dataOriginal histogram is

g a Two standard deviat ions stretch

o standard deviations stretch uses the data that are between d +2 standard deviations from the mean of the file values and hes them to the complete range of output screen values.

Output data

Input data

Using a Min/max contrast st retch

A Min/max contrast stretch makes the range of thlinearly from the minimum statistics value to thestatistics value in the input direction, and from 0brightness value in the output direction.

Input data

Output data

VISUA 127

Usin

A Licontrimag

a narrow range—vice is capable of total range of the rence in images

ch.

orm without any

utomatically the Stereo s not on and after e. This option loating Cursor is he 3D Floating Cursor Mode.

ce may decrease. lygon features is

nd not the texture Polygon Outlines re of the polygon. lygon features in

eatures are filled, ncerned with the s in a large area are acceptable in

LIZING IN STEREO

Input dataOriginal histogram is grey

If a large feature dataset is being edited, performanOne option for increasing the display speed of podisplaying only the outline of polygon features a(fill) of the polygon itself. Selecting the Display Only option displays the polygon but not the textuThis option is only applicable to the display of pothe Stereo Window and not the ArcMap display.

If this check box isn’t checked, then all polygon fwhich masks the feature below it. If you aren’t codetails of features beneath polygons (such as treeidentified as a parcel of land), then filled polygonsaddition to the outlines.

g a Linear contrast stretch

near contrast stretch is a simple way to improve the visible ast of an image. It is often necessary to contrast-stretch raw e data so that they can be seen on the display.


Input data

Output data

In most raw data, the data file values fall within usually a range much narrower than the display dedisplaying. The range can be expanded to use thedisplay device. Note that you can only see a diffegreater than 8 bits with this type of contrast stret

Using no stretch at a l l

If you select None, the data is displayed in raw fcontrast adjustment.

Recentering the stereo cursor

The Automatic recenter of stereo cursor option arecenters the 3D Floating Cursor to the center ofWindow. This occurs when Fixed Cursor Mode iyou have completed collecting or editing a featurhelps minimize uncertainty as to where the 3D Flocated and establishes a consistent position for tCursor while not forcing you to operate in Fixed

Displaying polygon outl ines

LYST FOR ARCGIS128

The tillustr

USING STEREO ANA

op illustration shows filled polygons representing houses. The bottom ation shows the same polygons in the same area, but unfilled.

129

What’s next?

VISUAL

In the nCursordifferen

IZING IN STEREO

ext chapter, you’ll learn all about what the 3D Floating is, how to adjust it, how to use it in conjunction with t modes, and some keyboard shortcuts.

LYST FOR ARCGIS
USING STEREO ANA130

131

6App

on of the earth’s ollect data. As a an be positioned Cursor can

at on, below, or int, but in this

erstanding the ed to collect and

Cursor modes rk with and indow. Plus, ages and collect

IN THIS C

• Using th

• AdjustinFloating

• Selectinoptions

• Using th

• ApplyingMode op

• Checkininformat

• Using th

• Using ke

6
lying the 3D Floating Cursor
Since Stereo Analyst for ArcGIS creates a 3D digital representatisurface using imagery, a standard cursor (2D) cannot be used to cresult, Stereo Analyst for ArcGIS uses a 3D Floating Cursor that cin all three dimensions (X, Y, and Z). This way, the 3D Floating properly rest on the feature that is being collected or edited.

This 3D Floating Cursor gets its name because the cursor can floabove a feature. The cursor can also be referred to as a ground podocument, the cursor is referred to as a 3D Floating Cursor. Undoperation of the 3D Floating Cursor is very important since it is usedit features reliably in the Stereo Window.

In this chapter, you’ll learn about using the different 3D Floatingsuch as the Terrain Following Mode. You’ll also learn how to wodetermine the accuracy of the 3D Floating Cursor in the Stereo Wkeyboard shortcuts are described that can help you quickly view imfeatures.

HAPTER

e 3D Floating Cursor

g the position of the 3D Cursor

g 3D Floating Cursor

e Terrain Following Mode

other Terrain Following tions

g accuracy of 3D ion

e 3D Floating Cursor

yboard shortcuts

LYST FOR ARCGIS

Using the 3D Float ing Cursorages reference a reo, the image must be at the ccur, the feature

ad along a rolling ating Cursor must n the surface of collected.

cursor commonly in stereo. The 3D sed in Stereo te 3D geographic

ating Cursor must ile viewing in e feature being s of the 3D

ight location. The adjusted using a TAC Systems TopoMouse, or

132

A 3D Ffor the image o

When isimplyon the cursors

This anaFloating

USING STEREO ANA

glyph image of an area in Laguna Beach, CA illustrates the 3D Cursor resting on top of a mountain ridge.

the Immersion SoftMouse.

loating Cursor consists of an independent cursor displayed left image and an independent cursor displayed for the right f an image pair.

mages are not viewed in stereo, the 3D Floating Cursor appears to be two separate cursors that may or may not rest same feature. However, when viewed in stereo, the two fuse to create the perception of a 3D Floating Cursor.

The left and right cursors for the left and right imlocation. When a feature is being collected in steposition of the cursor for the left and right imageexact same feature and location. If this does not ocannot be reliably collected. For example, if a rohill is being collected, the elevation of the 3D Flobe adjusted so that the 3D Floating Cursor rests othe road each time a point (vertex) for the road is

It is referred to as a 3D Floating Cursor since the floats above, below, or on a feature while viewingFloating Cursor is the primary measuring mark uAnalyst for ArcGIS to collect and measure accurainformation.

While collecting or editing of features, the 3D Flobe positioned on the feature being collected. Whstereo, the 3D Floating Cursor is positioned on thcollected by adjusting the X, Y, and Z coordinateFloating Cursor until it is perceived to be at the rX, Y, and Z position of the 3D Floating Cursor isdigitizing device such as the system mouse, the IMouse-Trak Professional, the Leica Geosystems

133

Adjust ing the posi t ion of the 3D Float ing Cursor

rcGIS with the errain Following

cation of the 3D tice how the left continuously

hic area (this is Stereo Window).

reen colorblock.

APPLYI

The podigitizican be can be elevatiodigitiziFloatinthe Ste

Every tare comcompuorienteelevatiolong as

Placiauto

Stereo automathe 3D the digapproaplace th

Gett

If theof intgoodCurso

NG THE 3D FLOATING CURSOR

ng the 3D Floating Cursor matically

Analyst for ArcGIS uses an approach referred to as ted terrain following to automatically adjust the position of Floating Cursor so that it rests on the feature of interest as itizing device is moved about in the Stereo Window. This ch uses digital image correlation techniques to accurately e 3D Floating Cursor on the feature of interest.

The correlated 3D Floating Cursor is indicated by the g

sition of the 3D Floating Cursor can be adjusted using a ng device. For example, a system mouse with a scroll wheel moved to adjust the X and Y location while the scroll wheel adjusted up or down to either increase or decrease the n (Z) of the 3D Floating Cursor. Therefore, using the

ng device, you actually have full control over the 3D g Cursor, and can position it accurately and reliably within reo Window.

ime the 3D Floating Cursor is adjusted, new 3D coordinates puted for the 3D Floating Cursor. 3D coordinates are

ted using the sensor model information associated with each d image in the image pair. It is important to note that an n model is not required to collect reliable 3D GIS data as

oriented images are being used.

ing the best accuracy

3D Floating Cursor is not accurately placed on the feature erest, the accuracy of the elevation of the feature won’t be . Manually adjusting the position of the 3D Floating r requires your continuous attention.

To observe this process, use Stereo Analyst for AStereo Window in the 3-Pane View and turn on TMode. Use the digitizing device to change the loFloating Cursor. While viewing in stereo, also noand right image positions of the cursors are beingadjusted so that they reference the same geograpobvious in the 2-Pane View at the bottom of the

LYST FOR ARCGIS134

Corrusedyou cWindpositand iimag

Usi

Stereuse iSteremenucategexist

Use ttoolba

customize is the these commands

re Editing.

e described in the

t ing Cursor

Cursor command h as meters.

USING STEREO ANA

he Customize dialog to add commands that do not already appear on rs.

following sections.

Decreasing elevat ion of the 3D Floa

Use the Decrease the elevation of the 3D Floatingto lower the 3D Floating Cursor by one unit, suc

elation is occurring automatically as the digitizing device is to modify the position of the 3D Floating Cursor. Make sure heck the colorblock located at the bottom right of the Stereo ow. If it is green, the 3D Floating Cursor is accurately

ioned; if it is red, the 3D Floating Cursor has failed to correlate s not resting on the same feature in both the right and left es of the image pair.

ng custom tools

o Analyst for ArcGIS provides you with some custom tools for n controlling the position of the 3D Floating Cursor in the o Window. You can access these tools by selecting the Tools , then Customize, then Commands. Click the Stereo Analyst ory to see the commands. You can drag the commands to any ing toolbar.

Another category that has tools you might want toLeica Feature Editing category. You can also dragto any existing toolbar.

The other category you can choose from is Leica Featu

Some notable Stereo Analyst for ArcGIS tools ar

APPL 135

Incr

Use tto rai

Rec

Use t3D F

Com

Use tfeatu

YING THE 3D FLOATING CURSOR

easing the elevat ion of the 3D Float ing Cursor

he Increase the elevation of the 3D Floating Cursor command se the 3D Floating Cursor by one unit, such as meters.

enter ing the 3D Float ing Cursor

he Recenter the 3D Floating Cursor command to reposition the loating Cursor in the middle of the Stereo Window.

plet ing feature col lect ion or edi t ing

he Complete editing feature command to finish collection of a re in the Stereo Window.

LYST FOR ARCGIS

Select ing 3D Float ing Cursor opt ionsize

d width, a point e images you are be larger or

the size of the 3D of the entire 3D

ne width

or adjusts the t equals a pixel.

hapes

the Stereo Floating Cursor can change the

g Cursor shapes.

136

You caoptimuthe Stelist, theFloatin

The 3D make chWindow

Adju

You mwhich ian optianaglypwhite.

USING STEREO ANA

Floating Cursor tab of the Stereo Analyst Options dialog is where you anges to the appearance of the 3D Floating Cursor in the Stereo .

sting 3D Floating Cursor color

ay want a different color display for the 3D Floating Cursor, s white by default. While viewing in quad-buffered stereo, mum 3D Floating Cursor color is red. While viewing in h, optimum 3D Floating Cursor colors are yellow and

n specify settings for the 3D Floating Cursor that provide m viewing for your application. To set these options, access reo Analyst toolbar, then click the Stereo Analyst dropdown n choose Options. Click the 3D Floating Cursor tab. The 3D g Cursor color, size, width, and shape can be modified there.

Adjusting 3D Floating Cursor s

When referring to the 3D Floating Cursor size anequals a pixel. Depending on the resolution of thworking with, you may find that the size needs tosmaller. An optimum size is 28 points. AdjustingFloating Cursor refers to adjusting the linear sizeFloating Cursor.

Adjusting 3D Floating Cursor l i

Adjusting the line width of the 3D Floating Cursthickness of the lines used to construct it. A poin

Adjusting 3D Floating Cursor s

By default, the 3D Floating Cursor shape used inWindow is a shaped like a cross (+). Certain 3D shapes are better for different applications—youshape quickly to see which works best for you.

The following table lists the different 3D Floatin

APPL 137

App

The Sand Znot a

3

loating Cursor is tivate hyperlinks perlink button on the hyperlink in


lying the 3D Floating Cursor in ArcMap

tereo Analyst for ArcGIS 3D Floating Cursor returns X, Y, coordinate information to any active ArcMap tool. Note that

ll ArcMap tools make use of the Z coordinate, however.

D Floating Cursor Shape 3D Floating Cursor Name

Dot

Cross

Open Cross

Open Cross with Dot

X

Open X

Open X with Dot

For example, the Stereo Analyst for ArcGIS 3D Falso able, like the cursor you see in ArcMap, to acassociated with features. You simply click the Hythe Tools menu, then click the feature to activatewhich you’re interested.

LYST FOR ARCGIS

Using the Terrain Fol lowing Mode

e is to specify an at TIN. In this

X, Y location of ion source at the lue.

lect the Constant tion for the scene.

ountainous an areas with

cceptable accuracy.

138

The Te3D Floto mantime yoTerrainStereo

The Tethat detFloatinsource

Usin

This mwhen aquality

In the se

USING STEREO ANA

tup above, the elevation source is a raster DEM file. Image correlation is set to 85 percent, which ensures a

rrain Following Mode can be used to automatically place the ating Cursor on the feature of interest so that you don’t have ually adjust the elevation of the 3D Floating Cursor every u collect a feature. You choose settings that control the Following Mode on the Terrain Following Cursor tab of the Analyst Options dialog.

rrain Following Mode provides two methods of operation ermine how elevation information is calculated for the 3D g Cursor. The options are either to use an external elevation or to use image correlation.

g external elevation information

ethod of determining 3D Floating Cursor elevation is useful dense elevation source is available, such as LIDAR, or if the of the Z coordinate is not of primary importance.

Using a Raster or TIN surface

One method used by the Terrain Following Modelevation source such as a DEM or an ESRI-formcase, Stereo Analyst for ArcGIS uses the currentthe 3D Floating Cursor and references the elevatsame X, Y location to determine the elevation va

Using a Constant e levat ion

If a raster or TIN file is not available, you can seelevation option, and then enter an average eleva

Using image correlation

This method works well in areas with rolling to mterrain. This method is less effective in dense urbshadow, or forested areas.

APPL 139

AnotFollopatchcorrefor Athe cmodeZ coo

For m“App

Usin

Withimagsensopointdirecsourc

UsinthemInstemay since

Usin

ThreperfocorreslopelocatOptio

patches (one from ) are compared, in value from 0 to fficient is 85.

es for a particular atch, Stereo sider statistically can be specified ining good match

alue increases the the correlation Minimum d and urban areas atch is high. A

reas where a ea of interest.

on can be more nd creates a n. Similarly, if ically different stressful to

the Terrain slope r ArcGIS to that the match of terest is flat with slider bar should


ad, elevation information comes strictly from the DEM, which be outdated due to construction, natural disaster, and so on it was created.

g the Correlat ion Opt ions

e correlation options are provided for optimizing the rmance of the Terrain Following Mode when image lation is used. These include correlation threshold, terrain specification, and image contrast specification, and are all ed on the Terrain Following Cursor tab of the Stereo Analyst ns dialog.

In images with a large amount of slope, correlatidifficult since the relief displacement on the grouparallax effect that increases with terrain variatioeach image of the image pair is collected at a radangle, the matching can be more computationallyprocess. Therefore, in both instances you can setslider bar to Steep. This forces Stereo Analyst foperform more extensive computations to ensure points between images is correct. If the area of invery little variation in elevation, the Terrain slopebe set to Flat.

her method (and the default method) used by the Terrain wing Mode is image correlation. In correlation, an image on the left image is used as a reference to search for the sponding image patch in the right image. Once Stereo Analyst rcGIS finds the matching area, correlation is achieved. Using orrelated left and right image positions along with the sensor l information, Stereo Analyst for ArcGIS calculates X, Y, and rdinates for the 3D Floating Cursor.

ore information about image correlation, refer to appendix C lying photogrammetry” on page 209.

g terrain fo l lowing with image correlat ion

image correlation, Stereo Analyst for ArcGIS consults the es themselves to derive 3D coordinate information. Using r model information and the correlated image positions of a on the ground, 3D coordinate information is computed tly from imagery without requiring an external elevation e.

g an external elevation source, like a DEM, the images selves are not consulted at all for elevation information.

Using Minimum correlat ion threshold

During the image correlation process, two image the left image and the other from the right imageand a correlation coefficient is computed ranging100. The optimum setting for the correlation coe

Since the correlation process finds multiple matchpoint on the ground in hope of finding the best mAnalyst for ArcGIS can be optimized to only convalid correlations. A correlation threshold value that weeds out possible false candidates while retacandidates.

Selecting a low Minimum correlation threshold vprobability of a false match, whereas increasing threshold may yield no correlation at all. A high correlation threshold value is preferred in foreste(with shadows) where the probability of a false mlow value is preferred in grassy areas and other aspecific land cover type is homogenous in the ar

Using Terrain s lope

LYST FOR ARCGIS140

Usin

In arpointcorreimagcorrecondSterecorre

Act

To acfollo

• CTd

• CV

• P

USING STEREO ANA

g Image contrast

eas of low contrast, it is more difficult to locate matching s in the left and right images of the image pair. Since the image lation process operates on the grey level values of the oriented es, the image contrast plays a vital role in the success of image lation. Setting the Image contrast level to reflect the actual ition of the imagery (that is, Low or High contrast) helps o Analyst for ArcGIS apply rigorous methods to improve the lation results.

ivating the Terrain Following Mode

tivate the Terrain Following Mode, you can do one of the wing:

lick the Apply continuous terrain following check box on the errain Following Cursor tab of the Stereo Analyst Options ialog.lick the Terrain Following Mode button located on the Stereo iew toolbar.ress the “t” key on the keyboard.

141

Apply ing other Terrain Fol lowing Mode opt ions

until you either Cursor tab, click wn below), or

.

more shortcut

rrain Following base ground ed.

n you are ted in the

APPLYI

You caFloatinWindothen cliThen, c

This is th

Usin

By clicspecifyMode. Mode oStereo


e Terrain Following Cursor tab of the Stereo Analyst Options dialog.

g continuous terrain fol lowing

king the Apply continuous terrain following check box, you that the 3D Floating Cursor is always in Terrain Following This way, you don’t have to toggle the Terrain Following n and off via the Terrain Following Mode button on the

View toolbar.

collecting telephone pole features. This is illustrafollowing diagram.

n choose other settings that control the behavior of the 3D g Cursor when it is in Terrain Following Mode in the Stereo w. To set these options, access the Stereo Analyst toolbar, ck the Stereo Analyst dropdown list and choose Options. lick on the Terrain Following Cursor tab. The Terrain Following Mode remains toggled on

uncheck the check box on the Terrain Followingthe recessed Terrain Following Mode button (shopress “t” on the keyboard to deactivate the mode

See “Using keyboard shortcuts” on page 149 forinformation.

Applying elevation bias

Elevation bias is used in conjunction with the TeMode. The Terrain Following Mode provides theelevation value to which the elevation bias is add

An example of when this feature is useful is whe

LYST FOR ARCGIS142

This iteleph

Regafeatu

1.

2.

3.

8 m

eter

s

oles, are ay be at 450 ill where the g the Terrain —the elevation by the Terrain

levation is added of the Stereo or vertex, and the

l lax

-parallax, you’ll fortable.

evice, such as the nside the Stereo e the 3D Floating the keyboard.

ove the mouse up elease the mouse

allax set to a

llax has been correction that is loating Cursor at

re information.

USING STEREO ANA

Mode and position it at the base of the feature. The elevation displays in the status bar at the bottom of the Stereo Window. For example, the base of the telephone pole feature may be at an elevation of 450 meters. Adjust the elevation of the 3D Floating Cursor so that it is at the top of the same feature and collect it. That elevation may add 8 additional meters, for a total elevation of 458 meters. Collect all remaining similar features at the top of the feature. Each individual base height as determined by the Terrain Following Mode plus the elevation bias, 8 meters, yields a total elevation for each separate feature.

and down to adjust the Y-parallax of the images. Rbutton and the “y” key when you have the Y-parcomfortable viewing level.

If Allow elevation bias is turned on, once Y-paraadjusted, Stereo Analyst for ArcGIS computes a applied to the elevation associated with the 3D Fthe time of collecting a feature.

See “Correcting Y-parallax” on page 203 for mo

llustration shows application of an elevation bias, 8 meters, to derive one pole feature height.

rding the diagram above, the steps to collect the telephone res are as follows:

Begin by putting the 3D Floating Cursor in Terrain Following

Note that not all features, in this case telephone pnecessarily at the same ground elevation—one mmeters, but the next is located on top of a small hground elevation is 451 meters. This is why usinFollowing Mode is important with elevation biasbias is added to the ground elevation determinedFollowing Mode.

When elevation bias is not in use, no additional eto features. The Z value you see in the status barWindow is the elevation recorded for that featurebias is set to zero (0).

Using elevat ion bias to af fect Y-para

If, while you are viewing in stereo, you perceive Ynotice that your perception of 3D may not be com

Y-parallax can be adjusted using the digitizing dsystem mouse. Position the 3D Floating Cursor iWindow (you may have to press the F3 key to givCursor focus), then press and hold the “y” key onThen, click and hold the left mouse button and m

APPL 143

Usi

An aModautompositcoordgrou

In ormustFloatcan sStere

SnapFloatare sdigitdoesthe 3utilit

Whilthe “grouare ckeyb

For mon pa

s

se it to position lso use Snap To op of buildings

ng Cursor over location in both ress the “s” key


e the 3D Floating Cursor is within the Stereo Window, press s” key on the keyboard to snap the 3D Floating Cursor to the nd. If you want to use the Snap To Ground feature while you ollecting features, you must apply it using the “s” key on the oard before you collect each vertex of the feature.

ore information about snapping in 3D, see “Using 3D Snap” ge 162.

ng Snap To Ground

lternative to continuously operating in the Terrain Following e is the Snap To Ground function. Snap To Ground

atically snaps the 3D Floating Cursor to an X, Y, Z ground ion. Snap To Ground consults the Z value at the X, Y inate you place the 3D Floating Cursor upon to determine the

nd elevation.

der for Snap To Ground to work, either an elevation source be specified, or image correlation is used to drive the 3D ing Cursor to the ground position at a particular location. You et these preferences on the Terrain Following Cursor tab of the o Analyst Options dialog.

To Ground works in both Fixed Image Mode (wherein the 3D ing Cursor moves freely in the Stereo Window, but the images tationary) and Fixed Cursor Mode, as well as while you are izing features. Note that the Snap To Ground functionality not work in conjunction with the Terrain Following Mode, as D Floating Cursor is already on the ground if you’re using that y.

Using Snap To Ground in other way

Although the name implies that you can only uthe 3D Floating Cursor on the terrain, you can aGround to position the 3D Floating Cursor on tand other features.

The process is the same: position the 3D Floatithe feature of interest in the same approximate the left and right image of the image pair, then pon the keyboard.

LYST FOR ARCGIS

Applying the Terrain Following Mode

USING STEREO ANA144

Before you begin this exercise, you should already have Stereo Analyst for ArcGIS installed, and both ArcMap and Stereo Analyst for ArcGIS should be running on your machine.

Using image correlation, the Terrain Following Mode checks pixels from both images of the image pair to automatically position the 3D Floating Cursor to rest on a feature of interest. All you have to do is position the 3D Floating Cursor in the Stereo Window in X and Y.

1. On the Stereo Analyst toolbar, click the Stereo Analyst dropdown list and choose Options.

2. On the Stereo Analyst Options dialog, click the Terrain Following Cursor tab.

3. Make sure that the Use image correlation option is active.4. Click OK on the Stereo Analyst Options dialog.

1

2

3 4

145

5. Click the Terrain Following Mode button on the Stereo 6 5

APPLYING THE 3D FLOATING CURSOR

Analyst toolbar.

When it is selected, it appears recessed on the toolbar.

6. Click the Fixed Cursor button on the Stereo Analyst Toolbar.

When it is selected, it appears recessed on the toolbar.

With the 3D Floating Cursor in a fixed location, you can use the X and Y thumb wheels at the right of the Stereo Window to adjust the displayed portion of the image pair. You can also use the Roam Tool to move around the image pair in the Stereo Window.

7. Move the X and Y thumb wheels up and/or down to adjust the position of the image pair “under” the 3D Floating Cursor to a desired position.

8. Note the elevation value displayed at the bottom of the Stereo Window, which updates as you adjust the image pair position.

As the 3D Floating Cursor moves to different positions, new X, Y, and Z coordinates are automatically computed. When image correlation succeeds, a green colorblock displays in the lower-right portion of the Stereo Window. When image correlation fails, a red colorblock displays in the same area.

7

8

Reading LE90 and CE90

In cases where the Terrain Following Mode fails to reliably position the 3D Floating Cursor on the ground or feature of interest, a red colorblock is displayed in the lower-right portion of the Stereo Window interface. See “Checking accuracy of 3D information” on page 146 for an explanation of the CE90 and LE90 computations.

LYST FOR ARCGIS

Checking accuracy of 3D informat ion work

and LE90 values o Window. These

lorblock, the e correlation has numbers indicate olorblock is only Following Mode.

the 3D Floating ature in both the

loating Cursor is re in both the left

146

If you cthe 3D CE90 athat areeach or

Both Cinformimagespositiois comp

CE90 athe 3D refers t3D coomeans to ± 1.7

The eq

The eq

In thesecoordin

USING STEREO ANA

uation to compute LE90 is as follows:

equations, sigma (Σ) represents the standard error of the ate in question.

CE90 ΣX ΣY+( )1.073=

LE90 ΣZ 1.646×=

heck the lower right-hand side of the Stereo Window when Floating Cursor is in Terrain Following Mode, you see nd LE90 with numerical values. These are accuracy indices derived from the sensor model information associated with iented image in an image pair.

E90 and LE90 are computed based on the sensor model ation that is part of the metadata associated with the oriented . This information is derived photogrammetrically when the n and attitude of the sensor as it existed as the time of capture uted.

nd LE90 provide a quality index for the current position of Floating Cursor. CE90 refers to circular error and LE90 o linear error. The 90 refers to the level of confidence in the rdinates of the point. For example, an LE90 of 1.765 meters that the current position of the 3D Floating Cursor is reliable 65 meters.

uation to compute CE90 is as follows:

Using CE90 and LE90 while you

As you work in Terrain Following Mode, CE90 are reported in the lower-right corner of the Sterevalues are theoretical error distribution values.

If the correlation has failed, indicated by a red conumbers are not representative of anything. If thsucceeded, indicated by the green colorblock, thethe standard deviation of the point feature. This cactive when the 3D Floating Cursor is in Terrain

If the colorblock is green, as shown below, then Cursor is correlated and is located on the same feleft image and the right image of the image pair.

If the colorblock is red, as shown below, the 3D Fnot correlated and is not located on the same featuimage and the right image of the image pair.

147

Using the 3D Float ing Cursorode is toggled off

rd Windows oolbars and from

w) as you move it use (if you don’t ystems y Toggle 3D eyboard.

Mode eliminates Floating Cursor

h the toggled on below.

tereo Window, it r does not appear. e of the Stereo

ables you to select ecting and editing rly helpful when features.

Toggle 3D

APPLYI

Togg

The Maon the Floatinyou preFloatin

Any adthe 3D Stereo to exit standar

Of coutoggledCursor

In this Cursorpositioparallayields aStereo Cursor


case, you are moving the images “beneath” the 3D Floating . Any adjustment of the mouse’s scroll wheel adjusts the n of the left and right images of the image pair, which affects x. Aligning the image pair so that the same features overlap ccurate elevation. Use the status bar at the bottom of the

Window to see the current elevation of the 3D Floating .

As you move the 3D Floating Cursor inside the Sis already active, and the standard Windows cursoAs you move the 3D Floating Cursor back outsidWindow, the standard cursor reappears, which endifferent target layers, editing modes, feature colltools, and the like. This functionality is particulayou use the ArcMap Editor to collect and/or edit

Some things to be aware of while using the AutoFloating Cursor Mode include:

l ing manually

nually Toggle 3D Floating Cursor Mode button is located Stereo View toolbar. When the Manually Toggle 3D g Cursor Mode button is toggled on (as depicted below) and ss the F3 key on the keyboard, you can freely move the 3D g Cursor and update its position in the Stereo Window.

justment of the mouse’s scroll wheel adjusts the elevation of Floating Cursor. Use the status bar at the bottom of the Window to see the current elevation. Press the F3 key again Manually Toggle 3D Floating Cursor Mode—you’ll see the d Windows cursor (an arrow) display in the Stereo Window.

rse, you might have the 3D Floating Cursor manually on in conjunction with Fixed Cursor Mode. The Fixed

Mode button, toggled on, is shown below.

When the Manually Toggle 3D Floating Cursor M(as depicted below), the mouse controls a standacursor, which allows you to make selections on tmenus.

You’ll see the standard Windows cursor (an arrointo the Stereo Window. This is the best mode tohave a special motion device like the Leica GeosTopoMouse). Remember, to reenter the ManuallFloating Cursor Mode, press the F3 key on the k

Toggling automatical ly

When active, the Auto Toggle 3D Floating Cursorthe need to press the F3 key in order to use the 3Din the Stereo Window. The button associated witAuto Toggle 3D Floating Cursor Mode is shown

LYST FOR ARCGIS148

• Yo“t

• Ycbta

• Yp(p

USING STEREO ANA

ou cannot roam over the entire image pair unless it is zoomed ut to a large extent, but you can use the keyboard shortcuts x” and “z” to adjust the scale of the image pair displayed in he Stereo Window.ou can’t use Auto Toggle 3D Floating Cursor Mode in

onjunction with Fixed Cursor Mode, wherein the images can e adjusted to improve the stereo view. The reason for this is hat the 3D Floating Cursor never exits the Stereo Window, nd therefore never is able to toggle off.ou can’t keep the 3D Floating Cursor positioned on a specific oint if you need to use the mouse for normal system events like making a selection from a menu) since you need to osition the cursor close to a window border to exit.

149

Using keyboard shortcuts

and Fixed Image out the image pair Mode and collect ode.

Mode and Fixed

tandard Windows options.

1.5 in the Stereo

lay by 1.5 in the

pair displayed so Stereo Window.

dges of the Stereo age 127 for more

APPLYI

Stereo that allyou timexecute

Usin

Press FWindothe 3D the Steclick incursor

Usin

Press Fthe Stethat it s

Usin

Press thshortcufollow Mode”

Usin

Press thwhen yto accuFloatinplaced Ground


t should be used when you want the 3D Floating Cursor to the terrain’s elevation. See “Using the Terrain Following on page 138 for more information.

g “s”

e “s” key to apply Snap To Ground. This shortcut is good ou’re doing feature extraction in an area where it is difficult rately place the 3D Floating Cursor on the ground. The 3D g Cursor’s elevation is automatically adjusted so that it is on the ground or feature of interest. See “Using Snap To ” on page 143 for more information.

Press the “x” key to zoom out of the area of dispStereo Window.

Using “r”

Press the “r” key to recenter the area of the imagethat the 3D Floating Cursor is in the middle of theThis shortcut is useful when navigating near the eWindow. See “Recentering the stereo cursor” on pinformation.

Analyst for ArcGIS includes a number of keyboard shortcuts ow you to quickly access common functions. They can save e since it isn’t necessary to move the 3D Floating Cursor to the command specified by the keyboard shortcut.

g F3

3 to toggle the 3D Floating Cursor on and off in the Stereo w. This switches between the regular, Windows cursor and Floating Cursor when the system mouse is positioned over reo Window and this shortcut is selected. You may have to side the Stereo Window before pressing F3 to give the focus.

g F4

4 to resynchronize the ArcMap display with the display in reo Window. This shortcut modifies the ArcMap display so hows the same geographic area as the Stereo Window.

g “t”

e “t” key to toggle the Terrain Following Mode. This

Using “c”

Press the “c” key to toggle between Fixed CursorMode. This can be useful if you prefer to roam abdisplayed in the Stereo Window in Fixed Image features in the Stereo Window in Fixed Cursor M

Using “i”

Press the “i” key to toggle between Fixed ImageCursor Mode. See “Using “c””, above.

Using “a”

Press the “a” key to activate the Arrow tool (the spointer), which can be used to select buttons and

Using “z”

Press the “z” key to zoom in the area of display byWindow.

Using “x”

LYST FOR ARCGIS150

Usin

Pressmustthen impreleva

USING STEREO ANA

g “y”

the “y” key to adjust Y-parallax. The 3D Floating Cursor be toggled on in the Stereo Window. Hold down the “y” key, hold down the left mouse button and move up and down to ove vertical separation. For more information, see “Using tion bias to affect Y-parallax” on page 142.

151

What’s next?

APPLYI

In the neditingusing toArcGIS


ext section, you can learn about feature collecting and in the Stereo Analyst for ArcGIS environment. You do so ols found in ArcMap and some new Stereo Analyst for tools.

LYST FOR ARCGIS
USING STEREO ANA152

Wor

ion 3

king with features

Sect

LYST FOR ARCGIS
USING STEREO ANA154

155

7Cap

f the existing ls are located on

ereo Window. ls to make your

e for feature ettings to collect

atures, and using

g devices. If you find it in the

IN THIS C

• Collectinmodes

• Using 3D

• Using S

• Using th

• Using d

7
turing GIS data
To collect features in Stereo Analyst for ArcGIS, you make use oArcGIS tools that you’re probably already familiar with. These toothe Editor toolbar, and can be applied both in ArcMap and the StStereo Analyst for ArcGIS also provides you with some new toofeature collection and editing easy in the Stereo Window.

In this chapter, you’ll learn about how to determine the best modcollection and whether or not you may be able to apply 3D Snap sadjacent 3D features.

You’ll also learn about applying Squaring settings to collect 3D fethe Monotonic Mode for special applications.

Finally, you’ll learn about the button mapping process for digitizinneed more detailed information about digitizing devices, you canStereo Analyst for ArcGIS On-line Help.

HAPTER

g features in different

Snap

quaring

e Monotonic Mode

igitizing devices

LYST FOR ARCGIS

Col lect ing features in di f ferent modes

es is appropriate in the Stereo the cursor focus, oating Cursor in

Mode” on page osition of the 3D rest without your ating Cursor. errain Following to be recessed, as

igitizing so that e of the CE90 and ich are located at dicate whether or cation to ensure

accuracy of 3D bout CE90 and

156

Stereo digitizedescrib

Usin

When tCursoryou macurrentMode, appears

Using Fwhen wand whthe featdisplay

You caAuto T

Usin

When tCursoraffect tmode isbest whWhen yon the follows

USING STEREO ANA

nnot use the Fixed Cursor Mode in conjunction with the oggle 3D Floating Cursor Mode.

g Fixed Image Mode

he Stereo Window is in Fixed Image Mode, the 3D Floating can move, but the images are fixed. Adjustments you make he separation and location of the 3D Floating Cursor. This appropriate when working with a system mouse, and works en the Auto Toggle 3D Floating Cursor Mode is in use. ou are in Fixed Image Mode, the Fixed Cursor Mode button

Stereo View toolbar does not appear to be recessed, as :

You can use the Terrain Following Mode while dthe 3D Floating Cursor is on the feature. Make notLE90 values and the red or green colorblock, whthe bottom, right of the Stereo Window. These innot the 3D Floating Cursor is correlated at that loaccuracy as you collect features. See “Checking information” on page 146 for more information aLE90.

Analyst for ArcGIS has different modes in which you can features in the Stereo Window. These different modes are ed in the following sections.

g Fixed Cursor Mode

he Stereo Window is in Fixed Cursor Mode, the 3D Floating is fixed in the center of the Stereo Window. Adjustments ke affect the position of the left image and right image of the ly displayed image pair. When you are in Fixed Cursor the Fixed Cursor Mode button on the Stereo View toolbar to be recessed, as follows:

ixed Cursor Mode while collecting features is appropriate orking in the Manually Toggle 3D Floating Cursor Mode

en you are using a TopoMouse. It is particularly useful when ure you’re digitizing extends beyond the Stereo Window .

Using Fixed Image Mode while collecting featurwhen the feature you’re digitizing fits easily withWindow. Click inside the Stereo Window to givethen press F3 on the keyboard to apply the 3D Flthe Stereo Window.

Using Terrain Fol lowing Mode

As you learned in “Using the Terrain Following 138, the Terrain Following Mode maintains the pFloating Cursor on the ground or a feature of intemanual adjustment of the elevation of the 3D FloWhen the Terrain Following Mode is active, the TMode button on the Stereo View toolbar appears follows:

CAPT 157

UsiMo

WhemoveWindactivWheToggStere

An aselecStere

Usi

As yaccesoptio

ding on the mode r ArcGIS are her tools are well ell as the On-line

URING GIS DATA

s to shortcuts by clicking the right mouse button. These ns are only available during feature collection and editing.

you’re in. The tools specific to Stereo Analyst foexplained in the rest of this chapter. All of the otdocumented in the book Editing in ArcMap as wHelp.

ng Auto Toggle 3D Floating Cursor de

n you are in Auto Toggle 3D Floating Cursor Mode, you can your 3D Floating Cursor freely inside and outside the Stereo ow without having to press the F3 key each time you want to

ate the 3D Floating Cursor for collecting or editing features. n you are in Auto Toggle 3D Floating Cursor Mode, the Auto le 3D Floating Cursor Mode button appears recessed in the o View toolbar, as follows:

dvantage to using this mode is that you can freely change your tions on the Editor toolbar, then move right back into the o Window to continue your work.

ng the context menu

ou collect and edit features in the Stereo Window, you have The options you’ll see on the menus change depen

LYST FOR ARCGIS

Collecting features in Fixed Image Mode

1

6

USING STEREO ANA158

The next steps tell you how to collect features in Fixed Image Mode in the Stereo Window. Fixed Image Mode is best used when the feature you want to collect displays wholly in the Stereo Window.

1. First, make sure that the editing toolbar is displayed by clicking the View menu, then pointing to Toolbars. Then, click Editor.

2. On the Stereo View toolbar, make sure that the Fixed Cursor Mode is not active. (If it is active, it appears recessed in the Stereo View toolbar.)

You cannot have the Fixed Cursor Mode active in conjunction with the Auto Toggle 3D Floating Cursor Mode. Click the button to deselect the Fixed Cursor Mode button if necessary.

3. On the Stereo View toolbar, click the Auto Toggle 3D Floating Cursor button.

4. On the Editor toolbar, click the Editor dropdown arrow and choose Start Editing.

5. On the Editor toolbar, make sure that the Task window displays Create New Feature.

6. On the Editor toolbar, click the Target dropdown list and choose the feature class into which you want the digitized feature to be stored.

7. On the Editor toolbar, click the Sketch Tool button.

At this point, when you move the cursor back into the Stereo Window, you’ll be able to digitize the feature of interest.

4

23

5

7

159

8. Click to collect the first vertex of the feature.

CAPTURING GIS DATA

9. Continue digitizing around the perimeter of the feature collecting corners.

10. Double-click to terminate collection of the feature.11. On the Editor toolbar, click the Editor dropdown list and

choose Stop Editing.12. Click Yes on the Save dialog to save the feature you just

collected, or No to discard the feature you just collected.

8

10

11

12

LYST FOR ARCGIS

Collecting features in Fixed Cursor Mode

1

USING STEREO ANA160

Another way to digitize features is in the Fixed Cursor Mode. In this mode, the feature’s position is adjusted by you “beneath” the 3D Floating Cursor. This mode is best used when the feature extends beyond the display of the Stereo Window.

1. First, make sure your Stereo Window is open and that the feature you want to digitize is clearly displayed with minimal parallax.

2. Make sure that the Editor toolbar is displayed. If it is not, click the Editor Toolbar button, or click the View menu, then point to Toolbars, then click Editor to display the toolbar.

3. On the Stereo View toolbar, click the Manually Toggle 3D Floating Cursor button and the Fixed Cursor Mode button. Both appear recessed.

4. On the Editor toolbar, click the Editor dropdown list and choose Start Editing.

2

3

4

Digi t iz ing features outs ide the d isplay

You may encounter a feature that extends outside the Stereo Window (depending on the scale at which you display the image pair). In a case like this, you can set an option so that the 3D Floating Cursor is recentered as you approach the extent of the Stereo Window. This way, you can continue to collect a feature that was initially extending beyond the Stereo Window.

From the Stereo Analyst toolbar, click the Stereo Analyst dropdown list, then click Options. Click the Stereo Display tab, then click Automatic Recenter for Stereo Cursor. Click Apply, then OK to activate the setting.

161

5. On the Editor toolbar, make sure that the Task window

6

9

CAPTURING GIS DATA

displays Create New Feature.6. On the Editor toolbar, click the Target dropdown list and

choose the appropriate feature class. 7. On the Editor toolbar, click the Sketch Tool button. 8. Move the cursor into the Stereo Window, click, then press

the F3 button on the keyboard to toggle on the 3D Floating Cursor.

You can tell that you are in Fixed Cursor Mode because as you move your mouse the image pair changes location in the Stereo Window, yet the 3D Floating Cursor remains in the center of the Stereo Window.

9. Click to collect the first vertex of the feature, then click to collect remaining vertices of the feature. When you are finished, double-click to complete the feature, or press F2 on the keyboard.

10. Press the F3 button on the keyboard to toggle off Manually Toggle 3D Floating Cursor Mode.

The Arrow Tool again appears in the Stereo Window, and the image pair no longer changes position with your mouse movement.

11. On the Editor toolbar, click the Editor dropdown list and choose Stop Editing.

12. Click Yes on the Save dialog to save the feature you just collected, or click No to discard the feature you just collected.

5

7

11

12

LYST FOR ARCGIS

Using 3D Snap

ith preferences ialog.

nce (indicated tolerance there red. The Y coordinates

162

You catoolbarthen clioptions

Usin

The setelevatio

The 3D elevatio

USING STEREO ANA

Snap tab is where you set tolerance values for coordinates in the Z, n, direction.

n change the settings for 3D Snap by accessing the Editor , then clicking the Editor dropdown list. Then click Options, ck the 3D Snap tab of the Editing Options dialog. These are specific to Stereo Analyst for ArcGIS.

g the sett ings on the 3D Snap tab

tings on the 3D Snap tab only control snapping in the Z, or n, direction.

Sett ing snapping in X and Y

The 3D Snap options also work in conjunction wset on the General tab of the Editing Options d

There, you see a preference for Snapping toleraby the red box, below). You can set the snappingas well as the units in which tolerance is measutolerance displayed in this window is for X andonly.

CAPT 163

Sett

If theSnapwell.set thconc

Sett

The CpositWindthe 3tolerof th

SettifrequIncrefrequ

App

The 3it mufeatusegm

On thbox fto 1 bCursto sn

osition of the 3D of 1 map unit; djacent roof snaps loating Cursor would not be

e map unit tolerance

bove, Stereo e 3D Snap tab: If

lect the vertex at the vertex would, nd Y coordinates. n snapping.

URING GIS DATA

lying 3D Snap tools

D Snap functionality controls the 3D Floating Cursor so that st be within a certain map unit range in order to snap to a re vertex (3D Vertex Snap) in X, Y, and Z; or part of a line ent (3D Edge Snap) in Z.

e 3D Snap tab of the Editing Options dialog, you see a check or Use elevation tolerance. The tolerance value, which is set y default, is measured in map units. That is, your 3D Floating

or must be within one map unit, such as a meter, in Z in order ap to the vertex.

elevation out of tolerance. Had you clicked to sethat location, either no snapping would occur, or indeed, snap to the existing vertex, but only in X aThe Z coordinate (elevation) would not be used i

Some 3D Snap keyboard shortcuts include:

• Endpoint—Ctrl + F5

• Vertex—Ctrl + F6• Midpoint—Ctrl + F7• Edge—Ctrl + F8

ing the snapping to lerance in Z

vertex you collect is not within the tolerance you set, the 3D tab on the Editing Options dialog has options for that case as If you are very interested in the accuracy of the Z coordinate, e preference so that no snapping occurs. If you are mostly

erned with the X and Y coordinates, click the 2D snap option.

ing cache s ize

ache size is used to store all features around the cursor ion as the 3D Floating Cursor moves around in the Stereo ow. The default cache size of 10 means that the cache around

D Floating Cursor covers a range 10 times the planar snapping ance. The planar snapping tolerance is set on the General tab e Editor Options dialog.

ng a small value causes the cache to be refreshed more ently, but reduces the time searching for the snapped feature. asing the cache causes the cache to be refreshed less ently, but increases the time searching for the snapped feature.

The following diagram illustrates snapping. The pFloating Cursor labeled 1 is within the tolerance therefore, clicking to select a vertex to begin the ato that vertex. However, the position of the 3D Flabeled 2 is outside the 1 map unit tolerance, andsnapped to the corner vertex.

In the above illustration, 3D Floating Cursor 1 is inside thand will be used for 3D Snap.

In the case of the 3D Floating Cursor labeled 2 aAnalyst for ArcGIS consults another setting on th

1

2

LYST FOR ARCGIS

Applying 3D snapping

USING STEREO ANA164

When you use 3D Snap, you are snapping to existing vertices in the X, Y, and Z direction.

1. You should already have Stereo Analyst for ArcGIS running and raster data and feature data displayed in the Stereo Window.

2. Click the Editor dropdown list, then choose Start Editing.3. Click the Editor dropdown list again, then choose

Snapping.

A window opens displaying the snapping agents.

4. In the window, click the plus sign next to Miscellaneous.5. Click the boxes to select either or both 3D Edge Snap

and 3D Vertex Snap.

3

2

4

5

165

6. Navigate in the Stereo Window to the feature to which

10

8

CAPTURING GIS DATA

you want to snap a new feature.7. On the Editor toolbar, make sure that the Task and

Target settings are correct.8. On the Editor toolbar, click the Sketch Tool.9. Position the 3D Floating Cursor near the existing feature.

Notice that a second, smaller cursor appears alongside the 3D Floating Cursor. This is to show you where the snapping is to occur.

10. Click to collect the first vertex.

Notice that the vertex is automatically snapped to either an edge or to a vertex, whichever is closest.

11. Continue to digitize your feature.12. Double-click to digitize the last vertex, or press F2 on the

keyboard, so that it also snaps to the existing feature.

12

LYST FOR ARCGIS

Customizing 3D Snap

5

USING STEREO ANA166

3D Snap can be easily accessed from a toolbar by customizing ArcMap.

1. In ArcMap, click the Tools menu, then click Customize.

This opens the Customize dialog.

2. On the Customize dialog, click the Command tab.3. Click the Leica Feature Editing category. 4. One by one, click, hold, and drag Enable Snap 3D Edge

and Enable Snap 3D Vertex to an existing toolbar, such as the Stereo View toolbar.

Once they are in place, selecting either of these buttons activates their respective functions, which is the same as checking the 3D snapping boxes in the Miscellaneous section of the Snapping window.

5. Click Close on the Customize dialog.

1

2

3 4

167

Using Squar inge digitizing quare a feature lerance value you

used in squaring cifies the greatest

ture.

an or equal to the moved. If the olerance to make

ssible. In the the ideal se the distance is in its original

ependent upon order or

CAPTUR

Squarinangles.polygo

You catoolbarthen cliclick thare spe

You set

ING GIS DATA

Squaring tolerances on the Square tab.

The tolerance value should be kept as small as pofollowing illustration, the red polygon representsrectangle. One of the vertices can be moved becauwithin the specified tolerance. The other remainslocation.

g is intended to help you rapidly collect features with right Squaring is only for use in digitizing polylines and ns—it is not applicable to point features.

n change the settings for Squaring by accessing the Editor , then clicking the Editor dropdown list. Then click Options, ck the Square tab. To use Squaring during feature collection, e check box labeled Enable squaring. The Squaring options cific to Stereo Analyst for ArcGIS.

Use of Squaring is most appropriate when you arfeatures with right angles. Squaring attempts to sbased on the rotation method you select and the tospecify.

Setting the tolerance

In each of the rotation modes, the tolerance valueis measured in map units. This tolerance value spedistance a vertex can be moved to square the fea

If a vertex can be moved by a value that is less thtolerance to make the feature corner square, it is vertex would have to be moved farther than the tthe corner square, it is not moved.

Predict ing resul ts

The results you get from Squaring are highly dthe tolerance setting, the rotation mode, and thedirection of digitizing.

LYST FOR ARCGIS168

Vertic

Set

You the fweiguses as alialignsides

Usin

WeigUsinweigalignvertipossi

The Wequa

surement the following sult of squaring.

ge rotation based

first and second feature.

examples is 10.0. .

s the first line original polygon;

USING STEREO ANA

hted mean is the default rotation mode used by Squaring. g the Weighted mean rotation mode means that the length-hted mean angle (R) of all sides is used to determine the ment. Once the alignment angle has been determined, the ces are adjusted within the tolerance to square corners where ble.

eighted mean function is expressed in the following tion:

Rra a×( ) rb b×( ) rc c×( ) rd d×( )+ + +

a b c d+ + +----------------------------------------------------------------------------------------------=

The red polygon represents the result of squaring

If you digitize clockwise, the First line rotation mode usedigitized as the basis for squaring. Figure A shows the Figure B shows the squared polygon.

1

2

3

4Figure A Figure B

es within tolerance are moved. Those outside tolerance aren’t moved.

t ing rotation mode

can choose from four methods to determine the alignment of eature. The first choice, Weighted mean, uses the length-hted angle of all sides to determine the alignment. First line the line formed by the first two digitized vertices of a feature gnment. Longest line uses the longest side of a feature as ment. Active view alignment makes the squared feature have either horizontal or vertical to the ArcMap data view.

g the Weighted mean rotat ion mode

Within tolerance

Outside tolerance

This method is good because it averages out meainaccuracies at all points. The tolerance value in example is 3.0. The red polygon represents the re

The Weighted mean rotation mode calculates the averaupon the length and angle of each segment.

Using the First l ine rotat ion mode

Using the First line rotation mode means that thevertices form the line which is used to square the

The tolerance value in each of the following two

a

ra

brb

crc

drd

CAPT 169

The Fdigitclockthe ficounbetw

Digitizvastly

Usin

Usingreatin wh

Di

Whdir

the feature as the Figure B shows the ed because it was

t ion mode

e borders of the a horizontal or

2

e the Orient air changes

p Display tab of

document to e “Orienting

URING GIS DATA

g the Longest l ine rotat ion mode

g the Longest line rotation mode means that the line with the est length is used to square the feature. In this mode, the order ich you digitize vertices does not matter.

git iz ing features

en using First line rotation, always digitize in a clockwise ection.

vertical direction, or both if possible.

Using Act ive v iew al ignment mode

When you are using this mode, it is best to havArcMap document to Image Pair when Image Poption on. This option is located on the ArcMathe Stereo Analyst Options dialog.

For more information about the Orient ArcMapImage Pair when Image Pair changes option, sedisplays” on page 119.

irst line rotation method is sensitive to the order in which you ize the feature vertices. If you digitize the feature in a wise direction, then the first line is the line formed between rst two vertices. However, if you digitize the feature in a ter-clockwise direction, then the first line is the line formed een the last two vertices.

ing the same line first but in the counter-clockwise direction can yield different results.

1

4

3

2

Figure A Figure B

The Longest line rotation mode uses the longest line ofbasis for squaring. Figure A shows the original polygon;squared polygon in red. NOTE: segment 4 was not movnot within the tolerance value.

Using the Act ive v iew al ignment rota

The Active view alignment rotation mode uses thArcMap data view to square the feature in either

1

3

4Figure A

Figure B

LYST FOR ARCGIS170

Whein theconsboth.Stere

The Aguide

Squ

Squamovethe listraigbe grverticreat

shows the squared E: Resultant line

2

2

2

USING STEREO ANA

aring polyl ines

ring polylines also considers making them straight. If greater ment of a vertex is required to make a right angle than to make ne straight, the lesser movement is taken and the line is made ht. As with squaring polygons, if movement of a vertex would eater than the tolerance value, the vertex is not moved. All ces are preserved, thus allowing you to edit the line after ing it.

n the Stereo Window and the ArcMap data view are oriented same direction, the results from this rotation mode are more

istent. That is, the final position of the feature is the same in Otherwise, the feature may appear in a different rotation in the o Window and the ArcMap data view.

ctive view alignment uses the ArcMap data view boundaries as a .

1

2

3

4Figure A

Figure B

Sides 1 & 3 adjustedto be parallelwith the ArcMapdata view.

Figure A shows the vertices outside tolerance. Figure Bpolyline. Figure C shows the straightened polyline. NOT(red) is shown slightly offset for clarity.

1Figure A

Figure C

1

1

Figure B

171

Configuring the Squaring tool1

3

CAPTURING GIS DATA

To set the proper options on the Square tab of the Editing Options dialog, please read “Using Squaring” on page 167.

1. On the Editor toolbar, click the Editor dropdown menu.2. Click Options to open the Editing Options dialog.3. On the Editing Options dialog, click the Square tab.4. Click the Enable squaring check box.5. Select the preferred Rotation mode from the dropdown

list.6. Set the Tolerance value.7. Click Apply, then OK.8. Collect features in the Stereo Window as usual.

2

4

5

6

7

LYST FOR ARCGIS

Using the Monotonic Mode

172

The Moensure either udesignefeature

The Modigitizior dowverticedeterm

Appl

1. Och

2. C3. R

Yex

4. Ave

5. D6. W

tode

USING STEREO ANA

rtex of the feature, and so on.ouble-click to complete the feature.hen you’ve finished, you can check the Sketch Properties see that the feature has been collected with increasing, creasing, or level elevation values, as appropriate.

notonic Mode is useful when you are collecting features to that the elevations of all points composing a feature run phill, downhill, or at the same elevation. This mode is d primarily for use in the collection of water drainage

s so that the water flows correctly, that is, not uphill.

notonic Mode bases its upward (in the case that you start ng at the water’s endpoint rather than starting point), same, nward flow on the elevation change between the first two s you collect. The increment rate of increase or decrease is ined by the 3D Floating Cursor elevation.

ying the Monotonic Mode

n the Editor toolbar, click the Editor dropdown list and oose Start Editing.

lick to select the first vertex of the feature.ight-click and select Monotonic from the context menu.

ou can also press the F10 key on the keyboard to enter and it Monotonic Mode.

djust the 3D Floating Cursor elevation to collect the next

3

6

173

Using digi t iz ing devices

igitizing devices.

uttons already er, you may want nique needs.

ach supported lyst for ArcGIS

pping buttons on

CAPTUR

You cacollect supporOn-linespecs.h

Addi

You’ll access dropdo

The Dev

Next, yattache

ING GIS DATA

ices dialog is your starting point for all device-related settings.

ou’ll specify the COM port to which the digitizing device is d in the Add Device dialog.

n use devices other than the standard system mouse to features in Stereo Analyst for ArcGIS. A list of the currently ted digitizing devices for Stereo Analyst for ArcGIS is in the Help and on the Web site <http://support.erdas.com/specs/tml>.

ng a digit izing device

use the Devices dialog to add a digitizing device. First, the Stereo Analyst toolbar, then click the Stereo Analyst wn list, then click Devices to open the dialog. The Add Device dialog gives you a list of the potential d

Mapping buttons

Each supported device comes with many of the bmapped to common digitizing functions. Howevto change the default settings to suit your own, u

A detailed list of optimum button mappings for edigitizing device can be found in the Stereo AnaOn-line Help.

The button mapping process is described in “Madigitizing devices” on page 174.

LYST FOR ARCGIS

Mapping buttons on digit izing devices

3

USING STEREO ANA174

In general, the procedure for mapping buttons on digitizing devices is as follows:

1. On the Stereo Analyst toolbar, click the Stereo Analyst dropdown list and choose Devices.

2. Click in the Device Selection window and select the digitizing device.

3. Click Button Mappings on the Devices dialog.4. To check a button mapping, click the button on the

digitizing device.

The button number you pressed appears in the window labeled Press/Select device button.

5. The currently assigned function (if there is one) appears in the Currently assigned to window.

1

2

4

5

175

6. To assign a new command to a button, select the button

9

8

11

12

CAPTURING GIS DATA

on the digitizing device.

Remember, you can use a combination of keys on some digitizing devices with the Shift button on the device. In the case of the TopoMouse, for example, that is button 4.

7. Select a function category, namely Stereo Analyst, from the Categories list.

Each category has its own list of commands. These commands display in the Commands/Buttons list.

8. Select the command you want to map to the selected button.

The current function displays in the Currently assigned to window.

9. Click Assign.10. Repeat steps 6 through 9 until you have mapped all of the

buttons on the device.11. Click Close on the Digitizing Device Button Mapping

dialog.12. Click Close on the Devices dialog.

For more information about digitizing devices and button mapping, see the Stereo Analyst for ArcGIS On-line Help.

7

LYST FOR ARCGIS

What’s next?

176

In the nyou aboused inphotogreferen

USING STEREO ANA

ext section, you’ll find the appendices. The appendices tell ut how data can be captured using imagery, how imagery is

stereo viewing, and how imagery is used in rammetry. There is also a glossary of terms for your ce.

App

ion 4

endices

Sect

LYST FOR ARCGIS
USING STEREO ANA178

179

ACap

e collection of ion and in a GIS are ta do not reflect

IN THIS A

• Collectin

• Preparin

• Using tr

• Applying

• Moving

• Identifyi

• Getting

A
turing data using imagery
This appendix gives you examples of how imagery is useful in thgeographic data. This data is of primary importance for the creatmaintenance of a GIS. If the data and information contained withinaccurate or outdated, the resulting analyses performed on the datrue, real-world applications and scenarios.

PPENDIX

g data for a GIS

g imagery for a GIS

aditional approaches

geographic imaging

from imagery to a 3D GIS

ng workflow

3D GIS data from imagery

LYST FOR ARCGIS

Col lect ing data for a GISthin the GIS re, collect, and er, GIS to face the

collect GIS data e can be said for

ow can the data

ect GIS data is the 1960 es?collect GIS data t. data is high. For ap an entire

does not include hardcopy maps is naccurate.sed to collect GIS ple, a building is d Y coordinate

eating DTMs, th’s geography to ected, the 3D eate a 3D GIS. time, cost, and nformation for a

gital mapping to and time. Also, s additional

180

Since itthe eartaccepteinformparcel opolygo

Variouas the f

• Ustopareaft

• Scresanscaqu

• ObgloareThass

• Oubuphaccim

• Apclaex

USING STEREO ANA

e resulting information is commonly merged into a GIS and ociated with existing vector datasets.tsourcing photogrammetric feature collection to service reaus. Traditional stereo plotters and digital otogrammetric workstations are used to collect highly urate geographic information such as orthorectified

agery, DTMs, and 3D vector datasets.plying remote sensing techniques, such as multispectral ssification, which traditionally have been used for tracting geographic information about the earth’s surface.

data provide only 2D information. For examrepresented with a polygon having only X aninformation. To create a 3D GIS involves crdigitizing contour lines, or surveying the earobtain 3D coordinate information. Once collinformation is merged with the 2D GIS to crEach approach is ineffective in terms of the accuracy associated with collecting the 3D i2D GIS.

• The cost associated with outsourcing core dispecialty shops is expensive in both dollars performing regular GIS data updates requireoutsourcing.

s inception and introduction, GIS was designed to represent h and its associated geography. Vector data has been d as the primary format for representing geographic

ation. For example, a road is represented with a line, and a f land is represented using a series of lines to form a

n.

s approaches have been used to collect the vector data used undamental building blocks of a GIS. These include:

ing a digitizing tablet to digitize features from cartographic, ographic, census, and survey maps. The resulting features stored as vectors. Feature attribution occurs either during or er feature collection.anning and georeferencing existing hardcopy maps. The ulting images are georeferenced and then used to digitize d collect geographic information. For example, this includes nning United States Geological Survey (USGS) 1:24,000

ad sheets and using them as the primary source for a GIS.taining ground surveying geographic information. Ground bal positioning system (GPS), total stations, and theodolites commonly used for recording the 3D locations of features.

These approaches have been widely accepted wiindustry as the primary techniques used to prepamaintain the data contained within a GIS; howevprofessionals throughout the world are beginningfollowing issues:

• The original sources of information used to are becoming obsolete and outdated. The samthe GIS data collected from these sources. Hand information in a GIS be updated?

• The accuracy of the source data used to collquestionable. For example, how accurate is topographic map used to digitize contour lin

• The amount of time required to prepare and from existing sources of information is grea

• The cost required to prepare and collect GISexample, georectifying 500 photographs to mcounty may take up to three months (which collecting the GIS data). Similarly, digitizingtime-consuming and costly, not to mention i

• Most of the original sources of information u

CAPT 181

Withthe ubecoback(incl

• R• G• O

Eachalthoin 2Dthe inapplidirec

Accur

URING DATA USING IMAGERY

ate 3D geographic information can be extracted from imagery.

the advent of image processing and remote sensing systems, se of imagery for collecting geographic information has me more frequent. Imagery was first used as a reference drop for collecting and editing geographic information uding vectors) for a GIS. This imagery included:

aw photography,eocorrected imagery, andrthorectified imagery.

type of imagery has its advantages and disadvantages, ugh each is limited to the collection of geographic information . To accurately represent the earth and its geography in a GIS, formation must be obtained directly in 3D, regardless of the

cation. Stereo Analyst for ArcGIS provides the solution for tly collecting 3D information from stereo imagery.

LYST FOR ARCGIS

Prepar ing imagery for a GIS

CPs,aph(s) using the , andwith the collected

format ion t ransparency

s directly, a during feature ver the the photographs. are recorded n has been

g steps are IS:

using a desktop es using a

res is then ormation is , rectified map,

CPs. Once the ic coordinates (X

ata is entered and features.

.

182

This seimagerassociafrom imsolutio

Usin

The folthe collimagerdigital not beethe imareferen

Examfrom

HardcoindustrForesteand urbdirectlyare comhardco

For theof photfield, inrecordehardcois recornonspa

Transfehardco

USING STEREO ANA

from hardcopy photographs. The hardcopy photographs monly used during fieldwork and research. As such, the

py photographs are a valuable source of information.

interpretation of 3D and height information, an adjacent set ographs is used together with a stereoscope. While in the formation and measurements collected on the ground are d directly onto the hardcopy photographs. Using the

py photographs, information regarding the feature of interest ded both spatially (geographic coordinates) and tially (text attribution).

rring the geographic information associated with the py photograph to a GIS involves the following steps:

scanner, or digitize only the collected featurdigitizing tablet.

• The resulting image or set of digitized featugeoreferenced to the earth’s surface. The infgeoreferenced to an existing vector coveragerectified image, or is georeferenced using Gfeatures have been georeferenced, geographand Y) are associated with each feature.

• In a GIS, the recorded tabular (attribution) dmerged with the digital set of georeferenced

This procedure is repeated for each transparency

ction describes the various techniques used to prepare y for a GIS. By understanding the processes and techniques ted with preparing and extracting geographic information agery, problem issues can be identified and the complete

n for collecting 3D geographic information can be provided.

g raw photography

lowing examples describe the common practices used for ection of geographic information from raw photographs and y. Raw imagery includes scanned hardcopy photography, camera imagery, videography, or satellite imagery that has n processed to establish a geometric relationship between gery and the earth. In this case, the images are not ced to a geographic projection or coordinate system.

ple 1: Col lect ing geographic informat ion hardcopy photography

py photographs are widely used by professionals in several ies as one of the primary sources of geographic information. rs, geologists, soil scientists, engineers, environmentalists, an planners routinely collect geographic information

• Scan the photograph(s),• Georeference the photograph using known G• Digitize the features recorded in the photogr

scanned photographs as a backdrop in a GIS• Merge and geolink the recorded tabular data

features in a GIS.

This procedure is repeated for each photograph.

Example 2: Col lect ing geographic infrom hardcopy photography using a

Rather than measure and mark on the photographtransparency is placed on top of the photographscollection. In this case, a stereoscope is placed ophotographs. Then, a transparency is placed overFeatures and information (spatial and nonspatial)directly on the transparency. Once the informatiorecorded, it is transferred to a GIS. The followincommonly used to transfer the information to a G

• Either digitally scan the entire transparency

CAPT 183

Exafrom

By sinterinforinfor

• G• I

t• I

t

This

App

Rawgeomterrameasnot binfor

Geopuse itechnestabgrourefergeog

Analyst for ArcGIS.

(or on approaches rtion or error iques have been applications,

narrow field of . General

can provide a little is known


mation are not reliable.

rocessing techniques warp, stretch, and rectify imagery for n the collection of 2D geographic information. These iques include geocorrection and orthorectification, which lish a geometric relationship between the imagery and the nd. The resulting 2D image sources are primarily used as ence backdrops or base image maps on which to digitize raphic information.

Conventional techniques of geometric correctiongeocorrection) such as rubber sheeting are basedthat do not directly account for the specific distosources associated with the imagery. These technsuccessful in the field of remote sensing and GISespecially when dealing with low resolution andview satellite imagery such as Landsat and SPOTfunctions have the advantage of simplicity. Theyreasonable geometric modeling alternative whenabout the geometric nature of the image data.

mple 3: Col lect ing geographic informat ion scanned photography

canning the raw photography, a digital record of the area of est becomes available and can be used to collect GIS mation. The following steps are commonly used to collect GIS mation from scanned photography:

eoreference the photograph using known GCPs.n a GIS, using the scanned photographs as a backdrop, digitize he features recorded on the photograph(s).n the GIS, merge and geolink the recorded tabular data with he collected features.

procedure is repeated for each photograph.

lying geoprocessing techniques

aerial photography and satellite imagery contain large etric distortion caused by camera or sensor orientation error,

in relief, earth curvature, film and scanning distortion, and urement errors. Measurements made on data sources that have een rectified for the purpose of collecting geographic

Spatial information can be collected easily using Stereo

Applying geocorrect ion

LYST FOR ARCGIS184

Prob

Convtime.imagdiffica readealisysteterrato ocmisabeingsuffe

Furthextraconvinfor

Solu

TechOrthtechnsourcOrthvery misashortgeogsolutprocedurin

agery have large tematic and

ues used in st efficiently, and m the raw

s of considering n between

e third dimension,

he ideal ographic as reference g GIS. Using d and then haracteristics. seamless

nly 2D geometric cted from stem. Collecting possible. The ndent on the

ects caused by the l source of input

M is another rchased, but at a

USING STEREO ANA

oBASE overcome all of these problems by using sophisticated iques to account for the various types of error in the input data es. This solution is integrated and accurate. IMAGINE

oBASE can process hundreds of images or photographs with few GCPs, while at the same time eliminating the lignment problem associated with creating image mosaics. In —less time, less money, less manual effort, and more raphic fidelity can be realized using the photogrammetric ion. Stereo Analyst for ArcGIS uses all of the information ssed in IMAGINE OrthoBASE and accounts for inaccuracies g 3D feature collection, editing, and interpretation.

Orthorectified images are limited to containing oinformation. Thus, geographic information colleorthorectified images is georeferenced to a 2D sy3D information directly from orthoimagery is imaccuracy of orthorectified imagery is highly depeaccuracy of the DTM used to model the terrain effearth’s surface. The DTM source is an additionaduring orthorectification. Acquiring a reliable DTcostly process. High-resolution DTMs can be pugreat expense.

lems

entional techniques generally process the images one at a They cannot provide an integrated solution for multiple es or photographs simultaneously and efficiently. It is very ult, if not impossible, for conventional techniques to achieve

sonable accuracy without a great number of GCPs when ng with high-resolution imagery, images with severe matic and/or nonsystematic errors, and images covering rough in such as mountain areas. Image misalignment is more likely cur when mosaicking separately rectified images. This lignment could result in inaccurate geographic information collected from the rectified images. As a result, the GIS rs.

ermore, it is impossible for geocorrection techniques to ct 3D information from imagery. There is no way for entional techniques to accurately derive geometric mation about the sensor that captured the imagery.

tion

niques used in Stereo Analyst for ArcGIS and IMAGINE

Orthorect i f icat ion

Geocorrected aerial photography and satellite imgeometric distortion that is caused by various sysnonsystematic factors. Photogrammetric techniqIMAGINE OrthoBASE eliminate these errors mocreate the most reliable and accurate imagery froimagery. IMAGINE OrthoBASE is unique in termthe image-forming geometry by using informatiooverlapping images and explicitly dealing with thwhich is elevation.

Orthorectified images, or orthoimages, serve as tinformation building blocks for collecting 2D geinformation required for a GIS. They can be usedimage backdrops to maintain or update an existindigitizing tools in a GIS, features can be collecteattributed to reflect their spatial and nonspatial cMultiple orthoimages can be mosaicked to form orthoimage base maps.

Problems

CAPT 185

Solu

StereinforUsinrequiinforinforis nothem


tion

o Analyst for ArcGIS allows for the collection of 3D mation—you are no longer limited to only 2D information. g sophisticated sensor modeling techniques, a DTM is not red as an input source for collecting accurate 3D geographic mation. As a result, the accuracy of the geographic mation collected in Stereo Analyst for ArcGIS is higher. There need to spend countless hours collecting DTMs and merging with your GIS.

LYST FOR ARCGIS

Using t radi t ional approaches the DTM? Many dited, since the vailable, or the

chniques such as to capture angles, e then required to n with the

e labor intensive, S technology. nk the 3D ing and merging

dditional errors to

action. Using two ints or a dispersed automatically merge the

contained in the

186

Unfortumeasurimagesmaps. Tsourcesgeorefe(heightfollowicollecte

Exam

The firtopograFurtherThe digmap or

Probl

The acccannot into yoto the q

Exam

The secgeogra

Probl

Where the origDTM iinto yo

USING STEREO ANA

uestionable scanning or digitization process.

ple 2

ond example involves merging existing DTMs with phic information contained in a GIS.

em

did the DTMs come from? How accurate are the DTMs? If inal source of the DTM is unknown, then the quality of the

s also unknown. As a result, any inaccuracies are translated ur GIS.

The next example involves automated DEM extroverlapping images, a regular grid of elevation ponumber of 3D mass points (that is, a TIN) can beextracted from imagery. You are then required toresulting DTM with the geographic information GIS.

nately, 3D geographic information cannot be directly ed or interpreted from geocorrected images, orthorectified , raw photography, or scanned topographic or cartographic he resulting geographic information collected from these is limited to 2D only, which consists of X and Y renced coordinates. In order to collect the additional Z ) coordinate, additional processing is required. The ng examples explain how 3D information is normally d for a GIS.

ple 1

st example involves digitizing hardcopy cartographic and phic maps and attributing the elevation of contour lines. interpolation of contour lines is required to create a DTM. itization of these sources includes either scanning the entire digitizing individual features from the maps.

em

uracy and reliability of the topographic or cartographic map be guaranteed. As a result, any error in the map is introduced ur GIS. Additionally, the magnitude of error is increased due

Can you easily edit and modify problem areas intimes, the problem areas in the DTM cannot be eoriginal imagery used to create the DTM is not aaccompanying software is not available.

Example 3

This example involves using ground surveying teground GPS, total stations, levels, and theodolitesdistances, slopes, and height information. You argeolink and merge the land surveying informatiogeographic information contained in the GIS.

Problem

Ground surveying techniques are accurate, but arcostly, and time-consuming—even with new GPAlso, additional work is required to merge and liinformation with the GIS. The process of geolinkthe 3D information with the GIS may introduce ayour GIS.

Example 4

CAPT 187

Pro

You For epositbuildrequisourcstep stepsmentyour

Exa

This colleprodphotocolleincluready

Pro

UsinrequiThe uphoto


blem

g these sophisticated and advanced tools, the procedures red for collecting 3D geographic information become costly. se of such equipment is generally limited to highly skilled grammetrists.

blem

are restricted to the collection of point elevation information. xample, using this approach, the slope of a line or the 3D ion of a road cannot be extracted. Similarly, a polygon of a ing cannot be directly collected. Many times, postediting is red to ensure the accuracy and reliability of the elevation es. Automated DEM extraction consists of just one required

to create the elevation or 3D information source. Additional of DTM interpolation and editing are also required, not to ion the additional process of merging the information with GIS.

mple 5

example involves outsourcing photogrammetric feature ction and data capture to photogrammetric service bureaus and uction shops. Using traditional stereoplotters and digital grammetric workstations, 3D geographic information is

cted from stereo models. The 3D geographic information may de DTMs, 3D features, and spatial and nonspatial attribution for input in your GIS database.

LYST FOR ARCGIS

Apply ing geographic imaging

ith transforming rmation. 3D ccurate or nd automated 3D GIS data can ographic imaging curate 3D need to digitize gital maps. These GIS data and ting a GIS with

188

To presrequireinform

• Acinf

• Co• Mi

co• Co

ha• Int

an

The onuse of irepresetypes osatellitemillimesatelliteimmed

Synthesensingthe fututechnolaccurat3D GISto as 3Dwill be

USING STEREO ANA

sizing the concepts associated with photogrammetry, remote , GIS, and 3D visualization introduces a new paradigm for re of digital mapping—one that integrates the respective ogies into a single, comprehensive environment for the e preparation of imagery and the collection and extraction of data and geographic information. This paradigm is referred geographic imaging. 3D geographic imaging techniques

used for building the 3D GIS of the future.

geographic imaging prevents the inclusion of inaoutdated information into a GIS. Sophisticated atechniques are used to ensure that highly accuratebe collected and maintained using imagery. 3D getechniques use a direct approach to collecting acgeographic information, thereby eliminating the from a secondary data source like hardcopy or dinew tools significantly improve the reliability ofreduce the steps and time associated with populaaccurate information.

erve the investment made in a GIS, a new approach is d for the collection and maintenance of geographic data and ation in a GIS. The approach must provide the ability to:

cess and use readily available, up-to-date sources of ormation for the collection of GIS data and information.llect accurate 2D and 3D GIS data from a variety of sources.nimize the time and cost associated with preparing, llecting, and editing GIS data.llect 3D GIS data directly from raw source data without ving to perform additional preparation tasks.egrate new sources of imagery easily for the maintenance d update of data and information in a GIS.

ly solution that can address all of those issues involves the magery. Imagery provides an up-to-date, highly accurate ntation of the earth and its associated geography. Various f imagery can be used, including aerial photography, imagery, digital camera imagery, videography, and 35 ter photography. With the advent of high-resolution imagery, GIS data can be updated accurately and

iately.

3D information can be used for GIS analysis.

3D geographic imaging is the process associated wimagery into GIS data or, more importantly, info

CAPT 189

The bphotomainphotospeciphotophotophotophoto

Overphotoand linto data

As a theirproddigitphotocolleautom

See amore


ppendix C, “Applying photogrammetry” on page 209 for information about photogrammetric applications.

ackbone of 3D geographic imaging is digital grammetry. Photogrammetry has established itself as the

technique for obtaining accurate 3D information from graphy and imagery. Traditional photogrammetry uses alized and expensive stereoscopic plotting equipment. Digital grammetry uses computer-based systems to process digital graphy or imagery. With the advent of digital grammetry, many of the processes associated with grammetry have been automated.

the last several decades, the idea of integrating grammetry and GIS has intimidated many people. The cost

earning curve associated with incorporating the technology a GIS has created a chasm between photogrammetry and GIS collection, production, and maintenance.

result, many GIS professionals have resorted to outsourcing digital mapping projects to specialty photogrammetric uction shops. Advancements in softcopy photogrammetry, or al photogrammetry, have broken down these barriers. Digital grammetric techniques bridge the gap between GIS data

ction and photogrammetry. This is made possible through the ated processes associated with digital photogrammetry.

LYST FOR ARCGIS

Moving f rom imagery to a 3D GIS

190

TransfocommoinformincludenonspaThrougautoma

Usin

Digitalof photGIS daphotogbreakthimagercamera

In ordesystemon the digitizecalibraphotogis not r

Convenmappinthe advbeen exphotog

Given tphotogused. Tcomprefile for

USING STEREO ANA

raphy to very high precisions. If high-end micron accuracy equired, more affordable desktop scanners can be used.

tional photogrammetric applications, such as topographic g and contour line collection, use aerial photography. With ent of digital photogrammetric systems, applications have tended to include the processing of oblique and terrestrial

raphy and imagery.

he use of computer hardware and software for rammetric processing, various image file formats can be hese include TIF, JPEG, GIF, Raw, Generic Binary, and ssed imagery, along with various software vendor-specific mats.

rming imagery into 3D GIS data involves several processes nly associated with digital photogrammetry. The data and

ation required for building and maintaining a 3D GIS orthorectified imagery, DTMs, 3D features, and the tial attribute information associated with the 3D features. h various processing steps, 3D GIS data can be tically collected and extracted from imagery.

g imagery

photogrammetric techniques are not restricted as to the type ography and imagery that can be used to collect accurate ta. Traditional applications of photogrammetry use aerial raphy (commonly 9 × 9 inches in size). Technological roughs in photogrammetry now allow for the use of satellite y, digital camera imagery, videography, and 35 millimeter photography.

r to use hardcopy photographs in a digital photogrammetric , the photographs must be scanned or digitized. Depending digital mapping project, various scanners can be used to photography. For highly accurate mapping projects,

ted photogrammetric scanners must be used to scan the

191

Ident i fy ing workf lownternal geometry tured. For aerial istortion, fiducial is normally rt. For digital ize of the charge-includes internal mber of columns l sensor model torical ed to determine ique is normally

tment and is

xact position and e imagery was inates. The

efined in terms of kappa (κ). Over

ce to collect S) information at s available, the y input for use in odel information an determine the

project using the

CPs in digital nd Z. The image le images. GCPs

rectified images, c maps.

CAPTUR

The wohierarcgeograwhich i

• De• Me• Co• Pe• Ex• Or• Co

This wrepeateexamplevery t

Defin

A sensoassociaand imcollectiassociabe knowtechniqcamerainforminform

ING DATA USING IMAGERY

ted with the camera or sensor used to capture photography agery. Since digital photogrammetry allows for the accurate on of 3D information from imagery, all of the characteristics ted with the camera/sensor, the image, and the ground must

n and determined. Photogrammetric sensor modeling ues define the specific information associated with a /sensor as it existed when the imagery was captured. This ation includes both internal and external sensor model ation.

external sensor model information can be directlphotogrammetric processing. If external sensor mis not available, most photogrammetric systems cexact position and orientation of each image in abundle block adjustment approach.

Measuring GCPs

Unlike traditional georectification techniques, Gphotogrammetry have three coordinates: X, Y, alocations of 3D GCPs are measured across multipcan be collected from existing vector files, orthoDTMs, and scanned topographic and cartographi

rkflow associated with creating 3D GIS data is linear. The hy of processes involved with creating highly accurate phic information can be broken down into several steps, nclude:

fine the sensor model,asure GCPs,llect tie points (automated),rform bundle block adjustment (that is, aerial triangulation),tract DTMs (automated),thorectify, andllect and attribute 3D features.

orkflow is generic and does not necessarily need to be d for every GIS data collection and maintenance project. For e, a bundle block adjustment does not need to be performed ime a 3D feature is collected from imagery.

ing the sensor model

r model describes the properties and characteristics

Internal sensor model information describes the iof the sensor as it exists when the imagery is capphotographs, this includes the focal length, lens dmark coordinates, and so forth. This informationprovided to you in the form of a calibration repocameras, this includes focal length and the pixel scoupled device (CCD) sensor. For satellites, this satellite information such as the pixel size, the nuin the sensor, and so forth. If some of the internainformation is not available (as in the case of hisphotography), sophisticated techniques can be usthe internal sensor model information. This technassociated with performing a bundle block adjusreferred to as self-calibration.

External sensor model information describes the eorientation of each image as they existed when thcollected. The position is defined using 3D coordorientation of an image at the time of capture is drotation about three axes: omega (ω), phi (ϕ), andthe last several years, it has been common practiairborne GPS and inertial navigation system (INthe time of image collection. If this information i

LYST FOR ARCGIS192

GCPaccusensothe bdata GCPphotousedoverl

Col

To pDTMthe ogrourecog

Tie ppointjoin relatimanusophcolleprepausedmult

App

Onceestabprojeproce

ation that is d 3D features, ocessing. The justment may ternal sensor and additional l. This output is rts outlining the xample, if the n is known, then ource data can be

int positions with ery, automated h’s surface using o as automated orrelation) measure the ithin the overlap

bundle block ints are tomated DTM f evenly eographic area of ated to create a any GIS f sight (LOS) al bedform on of

USING STEREO ANA

isticated, and automated techniques, tie points are now cted automatically, saving you time and money in the ration of 3D GIS data. Digital image matching techniques are

to automatically identify and measure tie points across iple overlapping images.

lying bundle block adjustment

GCPs and tie points have been collected, the process of lishing an accurate relationship between the images in a ct, the camera/sensor, and the ground can be performed. This ss is referred to as bundle block adjustment.

Using sensor model information determined fromadjustment, the image positions of the ground potransformed into 3D point positions. Once the auextraction process has been completed, a series odistributed 3D mass points is located within the ginterest. The 3D mass points can then be interpolTIN or a raster DEM. DTMs form the basis of mapplications including watershed analysis, line oanalysis, road and highway design, and geologicdiscrimination. DTMs are also vital for the creatiorthorectified images.

s serve a vital role in photogrammetry by establishing an rate geometric relationship between the images in a project, the r model, and the ground. This relationship is established using undle block adjustment approach. Once established, 3D GIS can be accurately collected from imagery. The number of s varies from project to project. For example, if a strip of five graphs is being processed, a minimum of three GCPs can be

. Optimally, five or six GCPs are distributed throughout the ap areas of the five photographs.

lecting t ie points

revent misaligned orthophoto mosaics and to ensure accurate s and 3D features, tie points are commonly measured within verlap areas of multiple images. A tie point is a point whose nd coordinates are not known; however, the tie point is visually nizable in the overlap area between multiple images.

oint collection is the process of identifying and measuring tie s across multiple overlapping images. Tie points are used to the images in a project so that they are positioned correctly ve to one another. Traditionally, tie points have been collected ally, two images at a time. With the advent of new,

Since it determines most of the necessary informrequired to create orthophotos, DTMs, DSMs, anbundle block adjustment is an essential part of prcomponents needed to perform a bundle block adinclude the internal sensor model information, exmodel information, the 3D coordinates of points,parameters (AP) characterizing the sensor modecommonly provided with detailed statistical repoaccuracy and precision of the derived data. For eaccuracy of the external sensor model informatiothe accuracy of 3D GIS data collected from this sdetermined.

Extracting DTMs

Rather than manually collecting individual 3D poa GPS or using direct 3D measurements on imagtechniques extract 3D representations of the eartthe overlap area of two images. This is referred tDTM extraction. Digital image matching (auto-ctechniques are used to automatically identify andpositions of common ground points appearing warea of two adjacent images.

CAPT 193

Ort

Orthinherinfortopogsystea GISan oras if servelayer

Col

3D Gto asinforalignviewfor thgeogprim

A 3Dinforadditcolle

Duridispl3D FearthGIS restsrests

nalyst for ArcGIS.

can be used to e ground so that the 3D Floating ple, the collection ated Terrain ting vertices. The

vector layers are haped to their n be transformed ographic imaging e attribute edited. Attribute collection of 3D


GIS allows for the direct collection of 3D geographic mation from a DSM using a 3D Floating Cursor. Thus, ional elevation data is not required. True 3D information is cted directly from imagery.

ng the collection of 3D GIS data, a 3D Floating Cursor is ayed within the DSM while viewing the imagery in stereo. The loating Cursor commonly floats above, below, or rests on the ’s surface or object of interest. To ensure the accuracy of 3D data, the height of the 3D Floating Cursor is adjusted so that it on the feature being collected. When the 3D Floating Cursor on the ground or feature, it can be accurately collected.

Cursor every time a feature is collected. For examof a feature in 3D is as simple as using the automFollowing Mode with the Sketch Tool, then collecresulting output is 3D GIS data.

For the update and maintenance of a GIS, existingcommonly superimposed on a DSM and then resaccurate real-world positions. 2D vector layers cainto 3D geographic information using most 3D gesystems. During the collection of 3D GIS data, thinformation associated with a vector layer can betables can be displayed with the DSM during theGIS data.

horectifying

orectification is the process of removing geometric errors ent within photography and imagery. Using sensor model mation and a DTM, errors associated with sensor orientation, raphic relief displacement, earth curvature, and other

matic errors are removed to create accurate imagery for use in . Measurements and geographic information collected from

thorectified image represent the corresponding measurements they were taken on the earth’s surface. Orthorectified images as the image backdrops for displaying and editing vector s.

lecting and attr ibuting 3D features

IS data and information can be collected from what is referred a digital stereo model (DSM). Based on sensor model mation, two overlapping images comprising a DSM can be ed, leveled, and scaled to produce a 3D stereo effect when ed with appropriate stereo viewing hardware. A DSM allows e interpretation, collection, and visualization of 3D

raphic information from imagery. The DSM is used as the ary data source for the collection of 3D GIS data.

Accurate 3D buildings can be extracted using Stereo A

Automated Terrain Following Mode capabilitiesautomatically place the 3D Floating Cursor on thyou do not have to manually adjust the height of

LYST FOR ARCGIS194

Interthe csuchAutodurinperimattribdurin

USING STEREO ANA

preting the DSM during the capture of 3D GIS data allows for ollection, maintenance, and input of nonspatial information as the type of tree and zoning designation in an urban area. mated attribution techniques simultaneously populate a GIS g the collection of 3D features with such data as area, eter, and elevation. Additional qualitative and quantitative ution information associated with a feature can be input g the collection process.

195

Gett ing 3D GIS data f rom imagery

stry applications.

restry companies amount of land, the amount e problems may areas.

CAPTUR

The protechniqfeaturewith a informedited. of 3D d

Find

The 3Dused fosimulatgeograforestryand tele

Appl

For fortree stacover),type offoreststgeograrequireDSM toarea.

Using 3can be Appropto delindirectly

ING DATA USING IMAGERY

and can be identified and measured as a 3D polygon. 3D phic imaging techniques are used to provide the GIS data d to determine the volume of a stand. This includes using a collect tree stand height, tree crown diameter, density, and

D DSMs with high-resolution imagery, various tree species identified based on height, color, texture, and crown shape. riate feature codes can be directly placed and georeferenced eate foreststand polygons. The feature code information is indexed to a GIS for subsequent analysis and modeling.

marketable timber located within a given plot of of timber lost due to fire or harvesting, and wherarise due to harvesting in unsuitable geographic

ducts resulting from using 3D geographic imaging ues include orthorectified imagery, DTMs, DSMs, 3D s, 3D measurements, and attribute information associated feature. Using these primary sources of geographic ation, additional GIS data can be collected, updated, and An increasing trend in the geocommunity involves the use ata in GIS spatial modeling and analysis.

ing 3D GIS applications

GIS data collected using 3D geographic imaging can be r spatial modeling, GIS analysis, and 3D visualization and ion applications. The following examples illustrate how 3D phic imaging techniques can be used for applications in , geology, local government, water resource management, communications.

ying 3D GIS to forestry

est inventory applications, an interpreter identifies different nds from one another based on height, density (crown species composition, and various modifiers such as slope, topography, and soil characteristics. Using a DSM, a

3D geographic imaging techniques can be used in fore

Based on the information collected from DSMs, fouse the 3D information in a GIS to determine the

LYST FOR ARCGIS196

App

Prioran insourcquanfromlithowith contr

DSMstrucDSMstruca DScan binforaccu

Appact

In orsourcresolup-toapplisupe

DSMfollo

entification and d land cover. lope, vegetation cal information, ed as 3D vectors.res soil mapping. nd collection of , soil texture, and of a given d.igh-resolution for various mportant.tal information t size, building y presence or other land use

fication and . Site selection ction, sanitary transmission line 3D topographic entory, land use,

phy collected extent of urban is categorized for rmine the extent

USING STEREO ANA

der to formulate social, economic, and cultural policies, GIS es must be timely, accurate, and cost-effective. High-ution imagery provides the primary data source for obtaining -date geographic information for local government cations. Existing GIS vector layers are commonly rimposed onto DSMs for immediate update and maintenance.

s created from high-resolution imagery are used for the wing applications:

landfill site selection, power plant siting, andlocation. Each application requires accurate representations, geologic inventory, soils invvegetation inventory, and so forth.

• Urban change detection studies use photografrom various time periods for analyzing the growth. Land use and land cover informationeach time period, and then compared to deteand nature of land use/land cover change.

lying 3D GIS to geology

to beginning expensive exploration projects, geologists take ventory of a geographic area using imagery as the primary e of information. DSMs are frequently used to improve the tity and quality of geologic information that can be interpreted imagery. Changes in topographic relief are often used in logical mapping applications since these changes, together the geomorphologic characteristics of the terrain, are olled by the underlying geology.

s are utilized for lithologic discrimination and geologic ture identification. Dip angles can be recorded directly on a in order to assist in identifying underlying geologic tures. By digitizing and collecting geologic information using M, the resulting geologic map is in a form and projection that e immediately used in a GIS. Together with multispectral mation, high-resolution imagery produces a wealth of highly rate 3D information for the geologist.

lying 3D GIS to local government ivit ies

• Land use/land cover mapping involves the idcategorization of urban and rural land use anUsing DSMs, 3D topographic information, stype, soil characteristics, underlying geologiand infrastructure information can be collect

• Land use suitability evaluation usually requiDSMs allow for the accurate interpretation asoil type, slope, soil suitability, soil moisturesurface roughness. As a result, the suitabilityinfrastructure development can be determine

• Population estimation requires accurate 3D himagery for determining the number of unitshousehold types. The height of buildings is i

• Housing quality studies require environmenderived from DSMs including house size, lodensity, street width and condition, drivewaabsence, vegetation quality, and proximity totypes.

• Site selection applications require the identiinventory of various geographic informationapplications include transportation route sele

CAPT 197

App

DSMand gfromwatecan bsurfichanchanrepre

Rathhighlestimwidthchandepoevolu

App

The ginfortelecare rbuildbuildOncecoveanalyinspe


mation for various applications associated with wireless ommunications. 3D geographic representations of buildings equired for radio engineering analysis and LOS between ing rooftops in urban and rural environments. Accurate 3D ing information is required to properly perform the analysis. the 3D data has been collected, it can be used for radio rage planning, system propagation prediction, plotting and sis, network optimization, antenna siting, and point-to-point ction for signal validation.

lying 3D GIS to resource management

s are a necessary asset for monitoring the quality, quantity, eographic distribution of water. The 3D information collected DSMs is used to provide descriptive and quantitative rshed information for a GIS. Various watershed characteristics e derived from DSMs including terrain type and extent,

cial geology, river or stream valley characteristics, river nel extent, river bed topography, and terraces. Individual river nel reaches can be delineated in 3D, providing an accurate sentation of a river.

er than manually survey 3D point information in the field, y accurate 3D information can be collected from DSMs to ate sediment storage, river channel width, and valley flat . Using historical photography, 3D measurements of a river

nel and bank can be used to estimate rates of bank erosion and sition, identify channel change, and describe channel tion and disturbance.

lying 3D GIS to telecommunications

rowing telecommunications industry requires accurate 3D

LYST FOR ARCGIS
USING STEREO ANA198

199

BUnd

about stereo llection.

IN THIS A

• Learningviewing

• Understparallax

• Understtranslati

• Underst

B
erstanding stereo viewing
This appendix provides you with detailed, technical information viewing and its effect on 3D stereoscopic viewing and feature co

PPENDIX

principles of stereo

anding stereo models and

anding scaling, on, and rotation

anding the epipolar line

LYST FOR ARCGIS

Learning pr incip les of stereo v iewingeo Analyst for f depth to include n.

s

ping images (an ptured from two simultaneously. ble depth

ated from the two

tion you achieve istance between independent

for 3D perception.

g the earth’s delineate map features are more y in stereo than in ows a stereo view.

200

Defin

On a dausing obinocumonocis refer

Using swith grbrain tophotogcreatina resultaccuractechniq

Stereo collecti

• Seelimooride

• Syimtria

• Thviesoupeinferr

USING STEREO ANA

entation information is required for the highly accurate termination of 3D information.stematic errors associated with raw photography and agery are considered and minimized during the block ngulation process.e collection of 3D coordinate information using stereo wing techniques is not dependent on a DEM as an input rce. Changes and variations in depth perception can be

rceived and automatically transformed using sensor model ormation and raw imagery. Therefore, DTMs containing or are not introduced into the collected GIS data.

These two overlapping photos can be viewed together

The importance of using images is that by viewinsurface in stereo, you can interpret, measure, andfeatures in 3D. The net benefit is that many map interpretable and have a higher degree of accurac2D with a single image. The following picture sh

ing stereoscopic viewing

ily basis, we unconsciously perceive and measure depth ur eyes. Persons using both eyes to view an object have lar vision. Persons using one eye to view an object have ular vision. The perception of depth through binocular vision red to as stereoscopic viewing.

tereoscopic viewing, depth information can be perceived eat detail and accuracy. Stereo viewing allows the human judge and perceive changes in depth and volume. In

rammetry, stereoscopic depth perception plays a vital role in g and viewing 3D representations of the earth’s surface. As , geographic information can be collected to a greater y in stereo as compared to traditional monoscopic ues.

feature collection techniques provide greater GIS data on and update accuracy for the following reasons:

nsor model information derived from block triangulation minates errors associated with the uncertainty of sensor del position and orientation. Accurate image position and

Digital photogrammetric techniques used in SterArcGIS extend the perception and interpretation othe measurement and collection of 3D informatio

Understanding how stereo work

A true stereo effect is achieved when two overlapimage pair), or photographs of a common area cadifferent vantage points, are rendered and viewedThe stereo effect, or ability to view with measuraperception, is provided by a parallax effect generdifferent acquisition points.

The stereo effect is analogous to the depth percepby looking at a feature with your two eyes. The dyour eyes represents two vantage points like twophotos, as in the following pictures.

UNDE 201

Using

Wheand tin deheighautomthe goverlwhicand tbetwby theleva

Duriobjecin thethe bvisuamove3D mcontiin ele

d to as a stereo ice relief, or ersion of a stereo

el information is ing an image pair. ing an image pair

on capabilities of ns in elevation t real-world 3D mple of a 3D hich is displayed

RSTANDING STEREO VIEWING

een the top and bottom of a building is perceived automatically e brain, and any changes in depth are due to changes in tion.

ng the stereo viewing process, the left eye concentrates on the t in the left image and the right eye concentrates on the object right image. As a result, a single 3D image is formed within

rain. The brain discerns height and variations in height by lly comparing the depths of various features. While the eyes across the overlap area of the two photographs, a continuous odel of the earth is formulated within the brain since the eyes nuously perceive the change in depth as a function of change vation.

This illustration shows a 3D model.

two images, a 3D stereo view is possible.

n viewing the features from two perspectives, (the left photo he right photo), the brain automatically perceives the variation pth between different objects and features as a difference in t. For example, while viewing a building in stereo, the brain atically compares the relative positions of the building and

round from the two different perspectives (that is, two apping images). The brain also determines which is closer and h is farther: the building or the ground. Thus, as the left eye he right eye view the overlap area of two images, depth

The 3D image formed by the brain is also referremodel. Once the stereo model is formed, you notvertical exaggeration, in the 3D model. A digital vmodel, a DSM, can be created when sensor modassociated with the left and right images comprisIn Stereo Analyst for ArcGIS, a DSM is formed usand accurate sensor model information.

Using the stereo viewing and 3D feature collectiStereo Analyst for ArcGIS, changes and variatioperceived by the brain can be translated to reflecinformation. The following picture shows an exafeature created using Stereo Analyst for ArcGIS, win IMAGINE VirtualGIS.

LYST FOR ARCGIS

Understanding stereo models and paral lax

positions of point A

aph (L1) at image rd motion of the wo ground points itions a' and b'. movement of larger than the d to X-parallax.

x associated with file view above, ound point B

ove (Pb).

'

202

Stereo pertainoverlappresentavailabthe preare two

Corr

The folpoints (GroundB is loc

The left different

The foland theground

USING STEREO ANA

and right Images of an image pair have the same features, but at locations.

lowing diagram illustrates a profile view of the image pair corresponding image positions of ground point A and point B.

Principal Point 1 Principal Point 2 The following diagram illustrates that the parallaground point A, depicted in the illustration of pro(Pa) is larger than the parallax associated with grdepicted in the illustration of the profile view ab

models provide a permanent record of 3D information ing to the given geographic area covered within the ping area of two images. Viewing a stereo model in stereo s an abundant amount of 3D information to you. The ility of 3D information in a stereo model is made possible by sence of what is referred to as stereoscopic parallax. There types of parallax: X-parallax and Y-parallax.

ecting X-parallax

lowing pictures show the image positions of two ground A and B) appearing in the overlapping area of two images. point A is located at the top of a building, and ground point ated on the ground.

A A

B B

This is a profile view of the image pair that illustrates theand point B.

Ground points A and B appear on the left photogrpositions a and b, respectively. Due to the forwaaircraft during photographic exposure, the same tappear on the right photograph (L2) at image posSince ground point A is at a higher elevation, theimage point a to position a' on the right image isimage movement of point b. This can be attribute

L1 L2

a bo

A

B

a' bo'

UNDE 203

This d

Thusgrouan imbrainvarioand heleva

UsinArcGassocheigh

ation.

e difficult. The reo viewing:

tographic scale between the eo view becomes

c collection. This rientation t, you experience DSM.rate sensor model llax between two

parallaxgher elevation260 meters)

parallaxwer elevation50 meters)


g 3D geographic imaging techniques, Stereo Analyst for IS translates and transforms the X-parallax information iated with features recorded by an image pair into quantitative t and elevation information.

left and right images. As a result, the 3D sterdistorted.

• Flight line misalignment during photographiresults in large differences in photographic obetween two overlapping images. As a resuleyestrain and discomfort while viewing the

• Erroneous sensor model information. Inaccuinformation creates large differences in paraimages comprising a DSM.

iagram illustrates parallax comparison between points.

, the amount of X-parallax is influenced by the elevation of a nd point. Since the degree of topographic relief varies across age pair, the amount of X-parallax also varies. In essence, the perceives the variation in parallax between the ground and us features, and therefore judges the variations in elevation eight. The following picture illustrates the difference in tion as a function of X-parallax.

Pa Pb

a' a b' b

xa' xbxa

xb'

o

Parallax changes with increases and decreases in elev

Correcting Y-parallax

Under certain conditions, viewing a DSM may bfollowing factors may influence the quality of ste

• Unequal flying height between adjacent phoexposures. This effect causes a difference in

X-Hi(~

X-Lo(~2X

Y

LYST FOR ARCGIS204

As a as Y-The famou

In this

The fY-pa

In this

e, translate, and ereo view is

erceived scale of e achieved by h image as

he differences in otographs were

he relative X and o minimize ons of the left and ng a flight line.

large relative a) for the left and

USING STEREO ANA

picture, Y-parallax doesn’t exist.

Y

X

result of these factors, the DSMs contain an effect referred to parallax, which introduces discomfort during stereo viewing. ollowing picture displays a stereo model with a considerable nt of Y-parallax.

picture, Y-parallax exists.

ollowing picture displays the same stereo model without rallax.

Y

X

To minimize Y-parallax, you are required to scalrotate the images until a clear and comfortable stavailable.

Scaling the stereo model involves adjusting the peach image comprising an image pair. This can badjusting the scale (that is, relative height) of eacrequired. Scaling the stereo model accounts for taltitude as they existed when the left and right phcaptured.

Translating the stereo model involves adjusting tY positions of the left and right images in order tX-parallax and Y-parallax. Translating the positiright images accounts for misaligned images alo

Rotating the left and right images adjusts for thevariation in orientation (that is, omega, phi, kappright images.

205

Understanding scal ing, t ranslat ion, and rotat ionr resampling or model

g display her accuracy.

UNDERS

When vthe leftrotatedit is youstereo vimagesorigina

When uAnalysimagerthe sterfor. ThreferredthroughparallaAnalysof the t

The folmodel

This DS

TANDING STEREO VIEWING

M doesn’t have sensor model information.

iewing a pair of tilted, overlapping photographs in stereo, and right images must be continually scaled, translated, and in order to maintain a clear, continuous stereo model. Thus, r responsibility to adjust Y-parallax in order to create a clear iew. Once properly oriented, you should notice that the

are oriented parallel to the direction of flight, which was lly used to capture the photography.

sing DSMs created from sensor model information, Stereo t for ArcGIS automatically rotates, scales, and translates the y to continually provide an optimum stereo view throughout eo model. Thus, the Y-parallax is automatically accounted e process of automatically creating a clear stereo view is to as epipolar resampling on the fly. As you roam out a DSM, the software accounts and adjusts for Y-

x automatically. Using OpenGL software technology, Stereo t for ArcGIS automatically accounts for the tilt and rotation wo images as they existed when the images were captured.

lowing picture displays a DSM created without sensor information.

The following picture displays the use of epipolatechniques for viewing a DSM created with sensinformation.

This DSM has sensor model information.

As a result of using automatic epipolar resamplintechniques, 3D GIS data can be collected to a hig

LYST FOR ARCGIS

Understanding the epipolar l ine

int to aid in the

with the states that the two und point, and the must all exist in a

x

Epipolar line

Exposure station 2

206

Geomemodel imagesmatchimatchi

The moassociaused todiagramlocated

Matchin

The folusing thshows two explines pkintersecconstraproblemproblemarea an

Image

the imafrom th

USING STEREO ANA

lowing diagram illustrates the image matching process e epipolar plane as a geometric constraint. The figure

the epipolar plane which is the plane that is defined by the osure stations ( L1 and L2 ) and the ground point, P. The and k′p′ are the epipolar lines and are defined by the tion of the images and the epipolar plane. Using epipolar int in the matching process transforms the matching from a two-dimensional problem to a one-dimensional , and is therefore beneficial since it reduces both the search

d the computation time (Wolf 1983).

The epipolar plane can be used as a geometric constraidentification of matching points.

Epipolar geometry is also commonly associated coplanarity condition. The coplanarity condition sensor exposure stations of an image pair, any grocorresponding image position on the two images common plane.

tric and radiometric characteristics (derived from sensor information and image grey values) associated with the comprising an image pair are used to constrain the image ng process in order to produce highly accurate and reliable ng image point pairs.

st common constraint, which is epipolar geometry ted with an image pair, is used to constrain the search area establish a pair of matching image points. The following illustrates an image point on a reference image being

along the epipolar line of an adjacent overlapping image.

g image points are located along the epipolar line.

point collected

Epipolar line

ge pair

Corresponding image

image pair

e left image of point located in theright image of the

L1 L2

z

y

YpZp

Xp

P

p p'

Epipolar plane

Source: Keating et al 1975

Exposure

k'k

station 1

UNDE 207

The cepipointera groalongdigitepipoconsof ea


ommon plane is also referred to as the epipolar plane. The lar plane intersects the left and right images, and the lines of

section are referred to as epipolar lines. The image positions of und point appearing on the left and right photos are located the epipolar line. The searching and matching process for

al image matching occurs along a straight line (that is, the lar line), thus simplifying the matching process. The epipolar

traint can only be applied if the image orientation and position ch sensor have been solved.

LYST FOR ARCGIS
USING STEREO ANA208

209

CApp

about IN THIS A

• Learningphotogr

• Acquirin

• Scannin

• Underst

• Understorientat

• Using d

C
lying photogrammetry
This appendix provides you with detailed, technical information photogrammetry, which is the foundation for stereo viewing.

PPENDIX

principles of ammetry

g images and data

g aerial photography

anding interior orientation

anding exterior ion

igital mapping solutions

LYST FOR ARCGIS

Learning pr incip les of photogrammetrygrammetry is to (such as photogrammetric llite images and topographic

aphic information nd elevation data. location of

s taken on the een objects using of plane table

easurement in he analog plotter, overlapping was topographic

210

Photoginformpicturebe view

This pic

Unde

PhotogreliablethroughphotogimagerPhotog

Photogcontinuphotogphotogphotogphotog

USING STEREO ANA

rammetry is the “art, science and technology of obtaining information about physical objects and the environment the process of recording, measuring and interpreting

raphic images and patterns of electromagnetic radiant y and other phenomena” (American Society of rammetry 1980).

rammetry was invented in 1851 by Aimé Laussedat, and has ed to develop since. Over time, the development of rammetry has passed through the phases of plane table rammetry, analog photogrammetry, analytical rammetry, and has now entered the phase of digital rammetry (Konecny 1994). This is an analog stereo plotter.

rammetric principles are used to extract topographic ation from aerial photographs and imagery. The following illustrates rugged topography. This type of topography can ed in 3D using Stereo Analyst for ArcGIS.

ture shows 3D topography.

rstanding photogrammetry

The traditional, and largest, application of photoextract topographic and planimetric information topographic maps) from aerial images. However,techniques have also been applied to process sateclose-range images to acquire topographic or noninformation about photographed objects. Topogrincludes spot height information, contour lines, aPlanimetric information includes the geographicbuildings, roads, rivers, and so on.

Prior to the invention of the airplane, photographground were used to extract the relationship betwgeometric principles. This was during the phase photogrammetry.

In analog photogrammetry, starting with stereo m1901, optical or mechanical instruments, such as twere used to reconstruct 3D geometry from two photographs. The main product during this phasemaps.

APPL 211

In anexpedevicanalydevephotoprod

Digitphotoare sscannManphotoortho

The oDEMTherWithphotoremo

DigitautomthereperfoIMAfollo

rpret information es the process of tellite imagery is mation from ative inating between

YING PHOTOGRAMMETRY

al photogrammetric systems employ sophisticated software to ate the tasks associated with conventional photogrammetry,

by minimizing the extent of manual interaction required to rm photogrammetric operations. One such application is GINE OrthoBASE, the interface of which is shown in the wing illustration.

called metric photogrammetry. Interpreting inforphotography and imagery is considered interpretphotogrammetry, such as identifying and discrimvarious tree types (Wolf 1983).

alytical photogrammetry, the computer replaced some nsive optical and mechanical components. The resulting es were analog/digital hybrids. Analytical aerotriangulation, tical plotters, and orthophoto projectors were the main

lopments during this phase. Outputs of analytical grammetry can be topographic maps, but can also be digital

ucts, such as digital maps and DEMs.

al photogrammetry, sometimes called softcopy grammetry, is photogrammetry applied to digital images that

tored and processed on a computer. Digital images can be ed from photographs or directly captured by digital cameras.

y photogrammetric tasks can be highly automated in digital grammetry (such as automatic DEM extraction and digital photo generation).

utput products are in digital form, such as digital maps, s, and digital orthophotos saved on computer storage media.

efore, they can be easily stored, managed, and used by you. the development of digital photogrammetry, grammetric techniques are more closely integrated into te sensing and GIS.

This is the IMAGINE OrthoBASE interface.

Photogrammetry can be used to measure and intefrom hardcopy photographs or images. Sometimmeasuring information from photography and sa

LYST FOR ARCGIS212

Ide

The taeriavertipointphotodowntopogcaptuillustresol

This i

Terrethe cand cappliengin

Obliqand ianglecommappli

hotographs or gital imagery can :

dsat, SPOT, and

e large geometric d nonsystematic se errors most for collecting grammetry is

g geometry, s, and explicitly

tion of the

nce to s sources. This

ital and video edium to large satellite

USING STEREO ANA

llustration shows a common satellite.

strial or ground-based photographs and images are taken with amera stationed on or close to the earth’s surface. Terrestrial lose-range photographs and images are commonly used for cations involved with archeology, geomorphology, civil eering, architecture, industry, and so on.

ue photographs and images are similar to aerial photographs mages, except the camera axis is intentionally inclined at an with the vertical. Oblique photographs and images are only used for reconnaissance and corridor mapping

cations.

distortion that is caused by various systematic anfactors. Photogrammetric processes eliminate theefficiently and provide the most reliable solutiongeographic information from raw imagery. Photounique in terms of considering the image-forminutilizing information between overlapping imagedealing with the third dimension: elevation.

Photogrammetric techniques allow for the collecfollowing geographic data:

• 3D GIS vectors• DTMs, which include TINs and DEMs

ntifying photographs and images

ypes of photographs and images that can be processed include l, terrestrial, close-range, and oblique. Aerial or vertical (near cal) photographs and images are taken from a high vantage above the earth’s surface. The camera axis of aerial or vertical graphy is commonly directed vertically (or near vertically) . Aerial photographs and images are commonly used for raphic and planimetric mapping projects and are commonly red from an aircraft or satellite. The following figure rates a satellite. Satellites use onboard cameras to collect high-ution images of the earth’s surface.

Digital photogrammetric systems use digitized pdigital images as the primary source of input. Dibe obtained from various sources. These include

• Digitizing existing hardcopy photographs,• Using digital cameras to record imagery,• Using sensors onboard satellites such as Lan

IRS to record imagery.

Using photogrammetry

Raw aerial photography and satellite imagery hav

Using terminology

This document uses the term imagery in referephotography and imagery obtained from variouincludes aerial and terrestrial photography, digcamera imagery, 35 millimeter photography, mformat photography, scanned photography, andimagery.

APPL 213

• O• D• T

In esgeogimagprocegrouareasphotoinforcollemain

YING PHOTOGRAMMETRY

rthorectified imagesSMsopographic contours

sence, photogrammetry produces accurate and precise raphic information from a wide range of photographs and es. Any measurement taken on a photogrammetrically ssed photograph or image reflects a measurement taken on the

nd. Rather than constantly go to the field to measure distances, , angles, and point positions on the earth’s surface, grammetric tools allow for the accurate collection of

mation from imagery. Photogrammetric approaches for cting geographic information save time and money, and tain the highest accuracies.

LYST FOR ARCGIS

Acquir ing images and data

orresponding sses the average e distance on the the flying height ith a flying height ers, the SI would

ed along a flight photos in the strip

e flying height tions. Camera tilt

combined to form consists of a of 20-30 percent.

ular block in e. The following

In cases where a r), photographic

mine SI, versus

214

Duringexposeapplicathan onimage,

Duringwhich timager

This illu

This illu

USING STEREO ANA

stration shows exposure stations in red over rough terrain.

stration shows exposure stations in blue along a flight path.

Flight pathof airplane

Exposure station

Flight Line 1

Flight Line 2

Flight Line 3

number of parallel strips, normally with a sidelap

A regular block of photos is commonly a rectangwhich the number of photos in each strip is the samillustration shows a block of 3 × 2 photographs. nonlinear feature is being mapped (such as a riveblocks are frequently irregular.

photograph or image collection, overlapping images are d along a direction of flight. Most photogrammetric tions involve the use of overlapping images. By using more e image, the geometry associated with the camera/sensor, and ground can be defined to greater accuracies.

the collection of imagery, each point in the flight path at he camera exposes the film, or the sensor captures the y, is called an exposure station.

Each photograph or image that is exposed has a cimage scale (SI) associated with it. The SI expreratio between a distance in the image and the samground. It is computed as focal length divided byabove the mean ground elevation. For example, wof 1000 meters and a focal length of 15 centimetbe 1:6667.

A strip of photographs consists of images capturline, normally with an overlap of 60 percent. All are assumed to be taken at approximately the samand with a constant distance between exposure starelative to the vertical is assumed to be minimal.

The photographs from several flight paths can be a block of photographs. A block of photographs

Determining SI

The flying height above ground is used to deterthe altitude above sea level.

APPL 215

This i

The f

In the

20sid

YING PHOTOGRAMMETRY

se illustrations, the area of overlap is indicated by the curly brackets.

llustration shows a regular rectangular block of aerial photos.

ollowing illustration shows two overlapping images.

Flying

Strip 2

Strip 1

60% overlap

-30% elap

direction

LYST FOR ARCGIS

Scanning aer ia l photographys uses, such as ote sensing ometric accuracy, units. The image matic tie point itive to scan errors can be ally derived.

erall accuracy of gery being used. scanning by the pixel

S data collection, appropriate he accuracy ect and the time

associated with .

216

Usin

Photogimage qscanneand ananecessaaccurac

These uboth inhave a micronresolutapprox

The reqAerial in the 115- to 3therefopixels. desiredresolut

Using

Desktodetail abut theyou showhich e

USING STEREO ANA

ions provides data with higher accuracy.

desktop scanners

p scanners are general-purpose devices. They lack the image nd geometric accuracy of photogrammetric-quality units, y are much less expensive. When using a desktop scanner, uld make sure that the active area is at least 9 × 9 inches, nables you to capture the entire photo frame.

The following table lists the scanning resolutionsvarious scales of photography and image file size

g photogrammetric scanners

rammetric scanners are special devices capable of high uality and excellent positional accuracy. Use of this type of

r results in geometric accuracies similar to traditional analog lytical photogrammetric instruments. These scanners are ry for digital photogrammetric applications that have high y requirements.

nits usually scan only film because film is superior to paper, terms of image detail and geometry. These units usually root mean square error (RMSE) positional accuracy of 4 s or less, and are capable of scanning at a maximum ion of 5 to 10 microns (5 microns is equivalent to imately 5,000 pixels per inch).

uired pixel resolution varies depending on the application. triangulation and feature collection applications often scan 0- to 15-micron range. Orthophoto applications often use 0-micron pixels. Color film is less sharp than panchromatic, re, color ortho applications often use 20- to 40-micron The optimum scanning resolution also depends on the photogrammetric output accuracy. Scanning at higher

Desktop scanners are appropriate for less rigoroudigital photogrammetry in support of GIS or remapplications. Calibrating these units improves gebut the results are still inferior to photogrammetriccorrelation techniques that are necessary for autocollection and elevation extraction are often sensquality. Therefore, errors attributable to scanningintroduced into GIS data that is photogrammetric

Choosing scanning resolutions

One of the primary factors contributing to the ov3D feature collection is the resolution of the imaImage resolution is commonly determined by theresolution (if film photography is being used), orresolution of the sensor.

In order to optimize the attainable accuracy of GIthe scanning resolution must be considered. The scanning resolution is determined by balancing trequirements versus the size of the mapping projrequired to process the project.

217

5 microns 0 dots per inch)

Ground Coverage(meters)

0.153

0.204

0.255

0.306

0.357

0.408

0.459

0.510

0.561

0.612

0.663

0.714

0.765

0.816

0.918

1.020

1.275

1.530

2.040

2.550

APPLYI
NG PHOTOGRAMMETRY
12 microns (2117 dots per inch)




8(30

Photo Scale1 to

Ground Coverage (meters)




1800 0.0216 0.0288 0.045 0.09

2400 0.0288 0.0384 0.060 0.12

3000 0.0360 0.0480 0.075 0.15

3600 0.0432 0.0576 0.090 0.18

4200 0.0504 0.0672 0.105 0.21

4800 0.0576 0.0768 0.120 0.24

5400 0.0648 0.0864 0.135 0.27

6000 0.0720 0.0960 0.150 0.30

6600 0.0792 0.1056 0.165 0.33

7200 0.0864 0.1152 0.180 0.36

7800 0.0936 0.1248 0.195 0.39

8400 0.1008 0.1344 0.210 0.42

9000 0.1080 0.1440 0.225 0.45

9600 0.1152 0.1536 0.240 0.48

10800 0.1296 0.1728 0.270 0.54

12000 0.1440 0.1920 0.300 0.60

15000 0.1800 0.2400 0.375 0.75

18000 0.2160 0.2880 0.450 0.90

24000 0.2880 0.3840 0.600 1.20

30000 0.3600 0.4800 0.750 1.50

LYST FOR ARCGIS

h scanned at 25 85 MB, assuming

imagery, the th the relationship

oordinate system he units in pixels, lumn and row

icrons ots per inch)

round verageeters)

.400

.250

.100

7

21

218

The Grmicrona squar

Unde

Concepimagermust be

Apply

The filewith itsas shownumbe

USING STEREO ANA

ound Coverage column refers to the ground coverage per pixel. Thus, a 1:40000 scale black and white photograps (1016 dots per inch) has a ground coverage per pixel of 1 meter × 1 meter. The resulting file size is approximatelye 9 × 9 inch photograph.

rstanding coordinate systems

tually, photogrammetry involves establishing the relationship between the camera or sensor used to capture the y itself, and the ground. In order to understand and define this relationship, each of the three variables associated wi defined with respect to a coordinate space and coordinate system.

ing a pixel coordinate system

coordinates of a digital image are defined in a pixel coordinate system. A pixel coordinate system is usually a c origin in the upper-left corner of the image, the x-axis pointing to the right, the y-axis pointing downward, and tn by axes c and r in the following illustration. These file coordinates (c, r) can also be thought of as the pixel co

rs, respectively.





85 m(300 d

Photo Scale1 to

Ground Coverage (meters)




GCo(m

40000 0.4800 0.6400 1.000 2.00 3

50000 0.6000 0.8000 1.250 2.50 4

60000 0.7200 0.9600 1.500 3.00 5

B/W File Size (MB) 363 204 84 21

Color File Size (MB) 1089 612 252 63

APPL 219

This ithe or

App

An imis usuplanecoordaeriaof opdiagrthe fimicr

system

wing illustration) ds a third axis (z). is defined at the lustration.

ce coordinate

system

X

A

YING PHOTOGRAMMETRY

ly ing an image coordinate system

age coordinate system or an image plane coordinate system ally defined as a 2D coordinate system occurring on the image with its origin at the image center. The origin of the image inate system is also referred to as the principal point. On

l photographs, the principal point is defined as the intersection posite fiducial marks as illustrated by axes x and y as in the am above. Image coordinates are used to describe positions on lm plane. Image coordinate units are usually millimeters or

ons. This illustration shows the image space and ground spasystems.

Z Height

Y

Ground coordinate system

llustration shows the origin of the image coordinate system (x, y) and igin of the pixel coordinate system (c, r).

y

x

c

r

Apply ing an image space coordinate

An image space coordinate system (see the follois identical to image coordinates, except that it adThe origin of the image space coordinate systemperspective center S as shown in the following il

z

y

x

Image coordinate

S

f

a

o

LYST FOR ARCGIS220

The pexistparalsysteimagto thusedin mimag

App

A grsystecoordvalueThis Z) in

Appcoo

Mostcurvavalueinclucoor

A geearthand tis pethreerule.

the center of the perpendicular at this center the local datum. ward, and the

is appendix does coordinates. d without adding

USING STEREO ANA

ture in their calculations. This is done by adding a correction or by computing geometry in a coordinate system that des curvature. Two such systems are the geocentric dinate system and the topocentric coordinate system.

ocentric coordinate system has its origin at the center of the ellipsoid. The Z-axis equals the rotational axis of the earth, he X-axis passes through the Greenwich meridian. The Y-axis rpendicular to both the Z-axis and X-axis, so as to create a -dimensional coordinate system that follows the right hand

erspective center is commonly the lens of the camera as it ed when the photograph was captured. Its x-axis and y-axis are lel to the x-axis and y-axis in the image plane coordinate m. The z-axis is the optical axis; therefore, the z value of an e point in the image space coordinate system is usually equal e focal length of the camera (f). Image space coordinates are to describe positions inside the camera, and usually use units illimeters or microns. This coordinate system is referenced as e space coordinates (x, y, z) in this appendix.

ly ing the ground coordinate system

ound coordinate system is usually defined as a 3D coordinate m that utilizes a known geographic map projection. Ground inates (X, Y, Z) are usually expressed in feet or meters. The Z is elevation above mean sea level for a given vertical datum. coordinate system is referenced as ground coordinates (X, Y, this appendix.

ly ing the geocentr ic and topocentr ic rd inate systems

photogrammetric applications account for the earth’s

A topocentric coordinate system has its origin atimage projected on the earth ellipsoid. The threecoordinate axes are defined on a tangential planepoint. The plane is called the reference plane or The X-axis is oriented eastward, the Y-axis northZ-axis is vertical to the reference plane (up).

For simplicity of presentation, the remainder of thnot explicitly reference geocentric or topocentricBasic photogrammetric principles can be presentethis additional level of complexity.

221

Using terrestrial photography

und space space.

a ground space

APPLYI

Photogcoordin

This illu

The imcoordin

NG PHOTOGRAMMETRY

rammetric applications associated with terrestrial or ground-based images utilize slightly different image and groate systems. The following figure illustrates the two coordinate systems associated with image space and ground

stration shows components associated with terrestrial photography.

age and ground space coordinate systems are right-handed coordinate systems. Most terrestrial applications use ate system defined using a localized Cartesian coordinate system.

ω

ϕ

κ

YG

XG

ZG

XA

ZAYA

Ground space

Ground point A

y

x

z

xa'a'

ya'

Perspective Center

Image space

ZL

YL

XL

Y

X

ϕ'

κ'

ω'

Z

LYST FOR ARCGIS222

The iimagsimilcoordexistpointcoord

Withκ (kagrouGCPangle

USING STEREO ANA

mage space coordinate system directs the z-axis toward the ed object and the y-axis directed north up. The image x-axis is ar to that used in aerial applications. The XL, YL, and ZL inates define the position of the perspective center as it

ed at the time of image capture. The coordinates of ground A (XA, YA, and ZA) are defined within the ground space inate system (XG, YG, and ZG).

this definition, three rotation angles ω (omega), ϕ (phi), and ppa) define the orientation of the image. You can also use the nd (X, Y, Z) coordinate system to directly define GCPs. Thus, s do not need to be transformed. Then the definition of rotation s ω', ϕ', and κ' is different, as shown in the figure on page 221.

223

Understanding inter ior or ientat ion

al length

he intersection of nter of the image erspective center

focal plane. For roximately 152

s, the focal length metric projects, determined

ge position where laboratory, this is llimation and

n the camera e the principal or any lens

with calculating ge position of the refore, the image e image, and then ducial mark.

APPLYI

Interiorsensor associadefininto transcoordinsystem

The folinternawhere Opoint.

This illu

The intfollowi

NG PHOTOGRAMMETRY

stration shows components of internal geometry.

ernal geometry of a camera is defined by specifying the ng variables:

x

Image plane

yo ya'

axa'

xo O

calibration report. Most applications prefer to uspoint of symmetry since it can best compensate fdistortion.

Defining fiducial marks

As stated previously, one of the steps associated interior orientation involves determining the imaprincipal point for each image in the project. Thepositions of the fiducial marks are measured on thcompared to the calibrated coordinates of each fi

orientation defines the internal geometry of a camera or as it existed at the time of image capture. The variables ted with image space are obtained during the process of g interior orientation. Interior orientation is primarily used form the image pixel coordinate system or other image ate measurement systems to the image space coordinate

.

lowing figure illustrates the variables associated with the l geometry of an image captured from an aerial camera,

represents the principal point and a represents an image

Perspective Center

y

z

Focal length Fiducial mark

• Principal point • Focal length• Fiducial marks• Lens distortion

Defining principal point and foc

The principal point is mathematically defined as tthe perpendicular line through the perspective ceplane. The length from the principal point to the pis called the focal length (Wang 1990).

The image plane is commonly referred to as the wide-angle aerial cameras, the focal length is appmillimeters, or 6 inches. For some digital camerais 28 millimeters. Prior to conducting photogramthe focal length of a metric camera is accurately (calibrated) in a laboratory environment.

The optical definition of principal point is the imathe optical axis intersects the image plane. In the calibrated in two forms: principal point of autocoprincipal point of symmetry, which can be seen i

LYST FOR ARCGIS224

Sincefor emarkcoord(rowcornerespebetwcoord

This dcoord

Usinpixeldefinusedcoordvalue

X

X

e calibrated of the measured ansformation en be used to el coordinates to

resented using an espondence and their rge RMSEs ted to film alibration

tion between the ge coordinate

e transformation rdinate system by an aerial ng procedure.

axis is referred to n also considers

nce between the affine

USING STEREO ANA

iagram shows the pixel coordinate system vs. the image space inate system.

g a 2D affine transformation, the relationship between the coordinate system and the image space coordinate system is ed. The following 2D affine transformation equations can be to determine the coefficients required to transform pixel inate measurements to the corresponding image coordinate s:

origin of the pixel coordinate system and the imasystem (xo-file and yo-file). Additionally, the affintakes into consideration rotation of the image cooconsidering angle Θ (theta). A scanned image ofphotograph is normally rotated due to the scanni

The degree of variation between the x-axis and y-as nonorthogonality. The 2D affine transformatiothe extent of nonorthogonality. The scale differex-axis and the y-axis is also considered using thetransformation.

the image space coordinate system has not yet been defined ach image, the measured image coordinates of the fiducial s are referenced to a pixel or file coordinate system. The pixel inate system has an x coordinate (column) and a y coordinate

). The origin of the pixel coordinate system is the upper left r of the image having a row and column value of 0 and 0, ctively. The following diagram illustrates the difference een the pixel coordinate system and the image space inate system.

Θ

Fiducial mark

Ya-file Yo-file

a-file

o-file

xa

a

ya

The x and y image coordinates associated with thfiducial marks and the X and Y pixel coordinatesfiducial marks are used to determine six affine trcoefficients. The resulting six coefficients can thtransform each set of row (y) and column (x) piximage coordinates.

The quality of the 2D affine transformation is repRMSE. The RMSE represents the degree of corrbetween the calibrated fiducial mark coordinatesrespective measured image coordinate values. Laindicate poor correspondence. This can be attribudeformation, poor scanning quality, out-of-date cinformation, or image mismeasurement.

The affine transformation also defines the transla

x a1 a2X a3Y+ +=

y b1 b2X b3Y+ +=

APPL 225

Def

Lenslocatexistdistotherepositthe d

This s

Radiradiadistocomm

Tangfromrepresmalnegliin a l

YING PHOTOGRAMMETRY

hows radial vs. tangential lens distortion.

al lens distortion causes imaged points to be distorted along l lines from the principal point o. The effect of radial lens rtion is represented as ∆r. Radial lens distortion is also

only referred to as symmetric lens distortion.

ential lens distortion occurs at right angles to the radial lines the principal point. The effect of tangential lens distortion is sented as ∆t. Because tangential lens distortion is much ler in magnitude than radial lens distortion, it is considered gible. The effects of lens distortion are commonly determined aboratory during the camera calibration procedure.

ining lens distor tion

distortion deteriorates the positional accuracy of image points ed on the image plane. Two types of radial lens distortion : radial lens distortion and tangential lens distortion. Lens rtion occurs when light rays passing through the lens are bent, by changing direction and intersecting the image plane at ions deviant from the norm. The following diagram illustrates ifference between radial and tangential lens distortion.

∆r ∆t

radial distance (r)

x

y

o

LYST FOR ARCGIS

Understanding exter ior or ientat ion

X

round point P

x´

226

Exteriothe campositioof exterthe chaor capt

The poZo. Therespectcommowhich i

The anthe relaand Z) rotationThey aillustraangles

USING STEREO ANA

These are elements of exterior orientation.

Y

Xp

Yp

Zp

Xo

Yo

r orientation defines the position and angular orientation of era that captured an image. The variables defining the

n and orientation of an image are referred to as the elements ior orientation. The elements of exterior orientation define racteristics associated with an image at the time of exposure ure.

sitional elements of exterior orientation include Xo, Yo, and y define the position of the perspective center (O) with to the ground space coordinate system (X, Y, and Z). Zo is nly referred to as the height of the camera above sea level, s commonly defined by a datum.

gular or rotational elements of exterior orientation describe tionship between the ground space coordinate system (X, Y, and the image space coordinate system (x, y, and z). Three angles are commonly used to define angular orientation.

re omega (ω), phi (ϕ), and kappa (κ). The following figures te the elements of exterior orientation and the individual (ω, ϕ, and κ) of exterior orientation.

Z

Zo

O

o p

G

ωκϕ

x

yz

y´

z´

yp

xp

f

APPL 227

The a

Omeabouphotocounrespeand d

The I(ISPRphotoThe xcoord

etween the image space coordinate

ined. A 3 × 3 systems is used. atrix, M. The

ential rotation of kappa about the

on

tween the camera grammetric tools ther.

e vector a can be to the image point e defined as the point P. The erring that a line point and to the

linear if one is a ing statement can

YING PHOTOGRAMMETRY

ga is a rotation about the photographic x-axis, phi is a rotation t the photographic y-axis, and kappa is a rotation about the graphic z-axis, which are defined as being positive if they are

terclockwise when viewed from the positive end of their ctive axis. Different conventions are used to define the order irection of the three rotation angles (Wang 1990).

nternational Society of Photogrammetry and Remote Sensing S) recommends the use of the ω, ϕ, κ convention. The graphic z-axis is equivalent to the optical axis (focal length). ', y', and z' coordinates are parallel to the ground space inate system.

utilize the following formulas in one form or ano

With reference to the figure on page 226, an imagdefined as the vector from the exposure station O p. A ground space or object space vector A can bvector from the exposure station O to the groundimage vector and ground vector are collinear, infextending from the exposure station to the imageground is linear.

The image vector and ground vector are only colscalar multiple of the other. Therefore, the followbe made:

ngles omega, phi, and kappa correspond to X, Y, and Z axes.

ϕ

κ

ω

z y

x

z y

x

z y

x

omega phi

kappa

Using the three rotation angles, the relationship bspace coordinate system (x, y, and z) and groundsystem (X, Y, and Z or x', y', and z') can be determmatrix defining the relationship between the twoThis is referred to as the orientation or rotation mrotation matrix can be defined as follows:

The rotation matrix is derived by applying a sequomega about the x-axis, phi about the y-axis, andz-axis.

Defining the coll inearity equati

The following section defines the relationship beor sensor, the image, and the ground. Most photo

Mm11 m12 m13

m21 m22 m23

m31 m32 m33

=

LYST FOR ARCGIS228

wherwithicomp

wherpoint

In orcoordrotat

wher

tween the station and age point location rity condition that collinearity e image, ground must all fall along ions comprise the

ground point n is commonly ra/sensor, the

13 Zp Zo1)–(

33 Zp Zo1–( )

----------------------------------

23 Zp Zo1)–(

33 Zp Zo1–( )

----------------------------------

USING STEREO ANA

der for the image and ground vectors to be within the same inate system, the ground vector must be multiplied by the

ion matrix M. The following equation can be formulated:

e

A Yp Yo–

Zp Zo–

=

a kMA= One set of equations can be formulated for each appearing on an image. The collinearity conditioused to define the relationship between the cameimage, and the ground.

yp yo– f–m21 Xp Xo1

)– m22 Yp Yo1)– m+(+(

m31 Xp Xo1–( ) m32 Yp Yo1

–( ) m+ +------------------------------------------------------------------------------------=

e k is a scalar multiple. The image and ground vectors must be n the same coordinate system. Therefore, image vector a is rised of the following components:

e xo and yo represent the image coordinates of the principal . Similarly, the ground vector can be formulated as follows:

a kA=

axp xo–

yp yo–

f–

=

Xp Xo–

The previous equation defines the relationship beperspective center of the camera/sensor exposureground point P appearing on an image with an imof p. This equation forms the basis of the collineais used in most photogrammetric operations. Thecondition specifies that the exposure station of thpoint, and its corresponding image point location a straight line, thereby being collinear. Two equatcollinearity condition:

xp xo–

yp yo–

f–

kMXp Xo–

Yp Yo–

Zp Zo–

=

xp xo– f–m11 Xp Xo1

)– m12 Yp Yo1)– m+(+(

m31 Xp Xo1–( ) m32 Yp Yo1

–( ) m+ +------------------------------------------------------------------------------------=

229

Using digi ta l mapping solut ions

used to determine th one image or ction uses the collinearity ure station, the ust be positioned

the X, Y, and Z to determine the h an image. Space ilable.

gle frame at a time. If echniques require rocessed.

the exterior riginating from at e through the

spective center of ment techniques, n can be lied to one image

APPLYI

Digitalorthorecreatiodeterm

For anybetweegroundthe rela

• Ex• Int• Ca

Well-kinteriorprojectlabor incontrolabundaparame

Dependsuch asblock aperformcreatioextensi

NG PHOTOGRAMMETRY

space resection, space forward intersection, and bundle djustment are used to define the variables required to orthorectification, automated DEM extraction, image pair

n, highly accurate point determination, and control point on.

image positions of the GCPs and resect at the perthe camera or sensor. Using least squares adjustthe most probable positions of exterior orientatiocomputed. Space resection techniques can be appor multiple images.

photogrammetry is used for many applications including ctification, automated elevation extraction, image pair n, stereo feature collection, highly accurate 3D point ination, and GCP extension.

of the aforementioned tasks to be undertaken, a relationship n the camera/sensor, the image(s) in a project, and the must be defined. The following variables are used to define tionship:

terior orientation parameterserior orientation parametersmera or sensor model information

nown obstacles in photogrammetry include defining the and exterior orientation parameters for each image in a using a minimum number of GCPs. Due to the costs and tensive procedures associated with collecting ground , most photogrammetric applications do not have an nt number of GCPs. Additionally, the exterior orientation ters associated with an image are normally unknown.

ing on the input data provided, photogrammetric techniques

Understanding space resection

Space resection is a technique that is commonly the exterior orientation parameters associated wimany images based on known GCPs. Space resecollinearity condition. Space resection using the condition specifies that, for any image, the exposground point, and its corresponding image point malong a straight line.

If a minimum number of three GCPs is known indirection, space resection techniques can be usedsix exterior orientation parameters associated witresection assumes that camera information is ava

Space resection is commonly used to perform sinorthorectification where one image is processed multiple images are being used, space resection ta minimum of three GCPs on each image being p

Using the collinearity condition, the positions oforientation parameters are computed. Light rays oleast three GCPs intersect through the image plan

LYST FOR ARCGIS

ints that appear in eters. The ough the the concept

230

Unde

Space fthe ovecollinecorrespassocia

This dia

USING STEREO ANA

rstanding space forward intersection

orward intersection is a technique that is commonly used to determine the ground coordinates X, Y, and Z of porlapping areas of two or more images based on known interior orientation and known exterior orientation paramarity condition is enforced, which states that the corresponding light rays from the two exposure stations pass thronding image points on the two images and intersect at the same ground point. The following diagram illustratested with space forward intersection.

gram illustrates space forward intersection.

X

Y

Z

Xp

Yp

Zp

Xo1 Yo2

Zo1

Xo2

Yo1

Zo2

O1

o1

O2

o2p1p2

Ground point P

APPL 231

SpacorienUsinparamp1 onYp, a

Spacappliusingdeter

Und

For mspaccan btheseorienimagphotoprovvalueaccu

Simithirtyminithe ticostl

The care lathe cTo enbund

mining the is computed ach image in a ts and adjusted s simultaneously known as least d solution for the ng error.

mathematical a block, the

e relationship has nformation collected in 3D.

ideography, and s commonly ocessing imagery lation is

The common ulation with the d the bundle ent is the most bution of errors. dition as the basis pace and ground

ith bundle block mages with e known. AGINE rates the

YING PHOTOGRAMMETRY

racies.

larly, rarely are there enough accurate GCPs for a project of or more images to perform space resection (that is, a

mum of 90 is required). In the case that there are enough GCPs, me required to identify and measure all of the points would be y.

osts associated with block triangulation and orthorectification rgely dependent on the number of GCPs used. To minimize

osts of a mapping project, fewer GCPs are collected and used. sure that high accuracies are attained, an approach known as

le block adjustment is used.

strip method, the independent model method, anmethod. Among them, the bundle block adjustmrigorous, considering the minimization and distriBundle block adjustment uses the collinearity confor formulating the relationship between image sspace.

In order to understand the concepts associated wadjustment, the following picture illustrates ten imultiple GCPs whose X, Y, and Z coordinates arAdditionally, six tie points are available. The IMOrthoBASE Graphic Status Display dialog illustphotogrammetric configuration.

e forward intersection techniques assume that the exterior tation parameters associated with the images are known. g the collinearity equations, the exterior orientation

eters along with the image coordinate measurements of point Image 1 and point p2 on Image 2 are input to compute the Xp,

nd Zp coordinates of ground point P.

e forward intersection techniques can also be used for cations associated with collecting GCPs, cadastral mapping airborne surveying techniques, and highly accurate point mination.

erstanding bundle block adjustment

apping projects having more than two images, the use of e intersection and space resection techniques is limited. This e attributed to the lack of information required to perform tasks. For example, it is fairly uncommon for the exterior tation parameters to be highly accurate for each photograph or e in a project since these values are generated grammetrically. Airborne GPS and INS techniques normally

ide initial approximations to exterior orientation, but the final s for these parameters must be adjusted to attain higher

A bundle block adjustment is best defined by exaindividual words in the term. A bundled solutionincluding the exterior orientation parameters of eblock and the X, Y, and Z coordinates of tie poinGCPs. A block of images contained in a project iprocessed in one solution. A statistical techniquesquares adjustment is used to estimate the bundleentire block while also minimizing and distributi

Block triangulation is the process of defining therelationship between the images contained withincamera or sensor model, and the ground. Once thbeen defined, accurate imagery and geographic iconcerning the earth’s surface can be created and

When processing frame camera, digital camera, vnonmetric camera imagery, block triangulation ireferred to as aerial triangulation (AT). When prcollected with a pushbroom sensor, block triangucommonly referred to as triangulation.

There are several models for block triangulation.models used in photogrammetry are block triang

LYST FOR ARCGIS
232
The pho

USING STEREO ANA

togrammetric block configuration displays in the IMAGINE OrthoBASE Graphic Status Display dialog.

GlossaryGlossary

This glossary defines terms commonly used in 3D GIS applications and photogrammetry.

to 2D images.

ges approximate

ponent is the

of approximately he amount of e of interest. You mouse. See also

ates elevation. A

but usually can be viewed in

Numerics

2D

Images or photos in X and Y coordinates only. There is no vertical element (Z)Viewed in mono, 2D images are good for qualitative analysis.

3D

Images or photos in X, Y, and Z (vertical) coordinates. Viewed in stereo, 3D imatrue earth features.

3D feature

A 3D feature is a feature that has vertex coordinates in X, Y, and Z. The Z comelevation of a particular vertex.

3D Floating Cursor

The 3D Floating Cursor is apparent when you have a DSM (that is, two images the same area) displayed. The 3D Floating Cursor’s position is determined by tX-parallax evident in the DSM and your positioning of it on the ground or featuradjust the position of the 3D Floating Cursor using the keyboard and the systemX-parallax.

3D model

A 3D model has vertex coordinates in X, Y, and Z, where the Z coordinate indic3D model displays in 3D (that is, a volumetric object).

Symbols

*.blk

An IMAGINE OrthoBASE block file. A block file can contain only one image,contains two or more images with approximately 60 percent overlap. Block files3D using Stereo Analyst for ArcGIS.

233

LYST FOR ARCGIS

*.img aerial triangulation

relationship he ground. The ation, DEM used when raphy, and

parameters to thogonality in een two

space coordinate

ns of exterior ation associated re. GPS provides on. See also

(as of finding the ps that frequently ebster OnLine

Interchange

e control codes, ccented letters puting 1999).

nted image pairs. ue glasses. These corresponding is produces a 3D

USING STEREO ANA234

An ERDAS IMAGINE image file. An .img file uses the hierarchical file format (HFA) structure to store many types of information in addition to the image data. For example, the .img format stores information about the file, sensor, layers, statistics, projection, and so on.

*.prj

A SOCET SET® project file, which contains sensor position and projection information about images in the project.

*.sup

A SOCET SET® support file, which contains geometric information about the image it supports.

κ

Kappa. The angle used to define angular orientation. Kappa is rotation about the Z-axis.

ω

Omega. An angle used to define angular orientation. Omega is rotation about the X-axis.

ϕ

Phi. An angle used to define angular rotation. Phi is rotation about the Y-axis.

Terms

additional parameter

(AP) In block triangulation, additional parameters characterize systematic error within the block of images and observations, such as lens distortion.

aerial photographs

Photographs taken from positions above the earth captured by aircraft. Photographs are used for planimetric mapping projects.

(AT) The process of establishing a mathematicalbetween images, a camera or sensor model, and tinformation derived is necessary for orthorectificgeneration, and image pair creation. This term isprocessing frame camera, digital camera, videognonmetric camera imagery.

affine transformation

A 2D plane-to-plane transformation that uses sixaccount for rotation, translation, scale, and nonorbetween the planes. Defines the relationship betwcoordinate systems such as a pixel and an image system.

airborne GPS

A technique used to provide initial approximatioorientation, which defines the position and orientwith an image as they existed during image captuthe X, Y, and Z coordinates of the exposure statiglobal positioning system.

algorithm

“A procedure for solving a mathematical problemgreatest common divisor) in a finite number of steinvolves repetition of an operation” (Merriam-WDictionary 2001).

American Standard Code for Information

(ASCII) A “basis of character sets...to convey somspace, numbers, most basic punctuation, and unaa–z and A–Z” (Free On-Line Dictionary of Com

anaglyph

A 3D image composed of two oriented or nonorieTo view an anaglyph, you require a pair of red/blglasses isolate your vision into two distinct partswith the left and right images of an image pair. Theffect with vertical information.

235

analog photogrammetry block file

e information ect, such as: or sensor model

etween imagery

s in a block file. ather, they are nt of overlap

exposures of a ock might consist –30 percent and

onship between nd. The ation, DEM

ach point ective center by a s for each image.

termines the sted at the time of s measured on the error , and GCPs. This

med on all

GLOSSARY

A technique in which optical or mechanical instruments such as analog plotters are used to reconstruct 3D geometry from two overlapping photographs.

analytical photogrammetry

A technique in which the computer replaces some expensive optical and mechanical components by substituting analog measurement and calculation with mathematical computation.

ASCII

See American Standard Code for Information Interchange.

AT

See aerial triangulation.

attribute

A piece of information about a feature, such as its length, location, name, and so on.

attribute table

A collection of attributes in tabular format.

attribution

Attribute data associated with a feature.

auto-correlation

A technique used to identify and measure the image positions appearing in the overlap of two adjacent images in a block file.

automated terrain following

Automatically places the 3D Floating Cursor on the ground so that you don’t have to manually edit the height of the 3D Floating Cursor. The X-parallax is adjusted using image correlation techniques to determine the image coordinate positions of a feature appearing on both the left and right image of an image pair.

binocular vision

Vision using two eyes. See also stereoscopic viewing.

A term used to describe and characterize all of thassociated with a photogrammetric mapping projprojection, spheroid, and datum; imagery; camerainformation; GCPs; and geometric relationship band the ground. A block file is a binary file.

block footprint

A graphical representation of the extent of imageThe images are not presented as raster images. Rdisplayed as vector outlines that depict the amoubetween images in the block file.

block of photographs

A series of photographs formed by the combinedflight. For example, a traditional frame camera blof a number of parallel strips with a sidelap of 20an overlap of 60 percent.

block triangulation

The process of establishing a mathematical relatiimages, the camera or sensor model, and the grouinformation derived is necessary for orthorectificgeneration, and image pair creation.

bundle

The unit of photogrammetric triangulation after emeasured in an image is connected with the perspstraight light ray. There is one bundle of light ray

bundle block adjustment

A mathematical technique (triangulation) that deposition and orientation of each image as they exiimage capture, determines the ground coordinateoverlap areas of multiple images, and minimizesassociated with the imagery, image measurementsis essentially a simultaneous triangulation perforobservations.

LYST FOR ARCGIS

calibration report collinearity condition

on, ground point, all be positioned

nother range, tching is often the range of data range of

ystem, expressed ) of the specified

ints. This round points on ates associated

etric techniques.

measure der to identify the erical or angular

ative orientation. timate five parameters ween two images

urposes of tie

USING STEREO ANA236

In aerial photography, the manufacturer of the camera specifies the interior orientation of each camera in the form of a certificate or report. Information includes focal length, principal point offset, radial lens distortion data, and fiducial mark coordinates.

Cartesian coordinate system

“A coordinate system consisting of intersecting straight lines called axes, in which the lines intersect at a common origin. Usually it is a 2-dimensional surface in which a ‘x, y’ coordinate defines each point location on the surface. The ‘x’ coordinate refers to the horizontal distance and the ‘y’ to vertical distance. Coordinates can be either positive or negative, depending on their relative position from the origin. In a 3-dimensional space, the system can also include a ‘z’ coordinate, representing height or depth. The relative measurement of distance, direction and area are constant throughout the surface of the system” (Natural Resources Canada 2001).

CCD

See charge-coupled device.

centroid

The point whose coordinates are the averages of the corresponding coordinates of the vertices of the polygon.

charge-coupled device

(CCD) A device in a digital camera that contains an array of cells that record the intensity associated with a ground feature or object.

coefficient

One number in a matrix, or a constant in a polynomial expression.

collinearity

A nonlinear mathematical model that photogrammetric triangulation is based upon. Collinearity equations describe the relationship among image coordinates, ground coordinates, and orientation parameters.

The condition that specifies that the exposure statiand its corresponding image point location must along a straight line.

contrast stretch

The process of reassigning a range of values to ausually employing a linear function. Contrast streused in displaying continuous raster layers since file values is commonly much narrower than thebrightness values available to the display device.

control point

A point with known coordinates in a coordinate sin the units (such as meters, feet, pixels, film unitscoordinate system.

control point extension

The process of converting tie points to control potechnique requires the manual measurement of gphotos of overlapping areas. The ground coordinwith GCPs are then determined using photogramm

coordinate system

“A system, based on mathematical rules, used tohorizontal and vertical distance on a surface, in orlocation of points by means of unique sets of numvalues” (Natural Resources Canada 2001).

coplanarity condition

The coplanarity condition is used to calculate relIt uses an iterative least squares adjustment to esparameters (By, Bz, omega, phi, and kappa). Theexplain the difference in position and rotation betmaking up the image pair.

correlation

Regions of separate images are matched for the ppoint or mass point collection.

237

correlation coefficient digital photogrammetry

t are stored and canned from l cameras.

s of digital pplications.

data array, (C, Y) and the

meric data, f vector data from traced using a

e on a display

as east to west). r satellite’s anner as the

of a sensor as it ctive center with

GLOSSARY

Represents the measure of similarity between a set of image points appearing within the overlapping portions of an image pair. A large correlation coefficient (0.80-1.0) statistically indicates that the set of image points is more similar than a set of image points which has a low correlation coefficient value (less than 0.50).

correlation threshold

A value used in image matching to determine whether to accept or discard match points. The threshold is an absolute value threshold ranging from 0.100 to 1.000.

cross-strips

Strips of image data that run perpendicular to strips collected along the flight line.

datum

“A datum is a system of reference for specifying the horizontal and vertical spatial positions of points” (Wolf and Dewitt 2000). See also reference plane.

DEM

See digital elevation model.

digital elevation model

(DEM) Continuous raster layers in which data file values represent elevation.

digital image matching

The process of matching features common to two or more images for the purpose of generating a 3D representation of the earth. Also known as auto-correlation.

digital orthophoto

An aerial photo or satellite scene that has been transformed by the orthogonal projection, yielding a map that is free of most significant geometric distortions.

Photogrammetry as applied to digital images thaprocessed on a computer. Digital images can be sphotographs or can be directly captured by digita

digital stereo model

(DSM) Stereo models that use imaging techniquephotogrammetry that can be viewed on desktop a

digital terrain model

(DTM) A discrete expression of topography in a consisting of a group of planimetric coordinates elevations of the ground points and breaklines.

digitizing

Any process that converts nondigital data into nuusually to be stored on a computer. The creation ohardcopy materials or raster images. The data aredigitizer keypad on a digitizing tablet, or a mousdevice.

direction of flight

The direction in which the craft is moving (such Images in a strip are captured along the aircraft odirection of flight. Images overlap in the same mdirection of flight.

draping

See feature draping.

DSM

See digital stereo model.

DTM

See digital terrain model.

elements of exterior orientation

Variables that define the position and orientationobtained an image. It is the position of the persperespect to the ground space coordinate system.

LYST FOR ARCGIS

elevation source feature collection

ing various types otely-sensed

nges in elevation

bjects on the ges.

e of an aerial cials are used to es to image

s at a constant in the display.

stant position display.

rplane consisting ptured along a

iple flight lines exposes the film. bined to form a

USING STEREO ANA238

Data, such as a DEM or TIN, that supplies the Z component of a feature or scene. An elevation source enables stereo viewing and 3D feature extraction.

ellipsoid

“A surface all plane sections of which are ellipses or circles” (Merriam-Webster OnLine Dictionary 2000).

endlap

The area common to two images in a strip of photos taken along the flight path. Another term for overlap. See also overlap.

epipolar line

The line traced on each image representing the intersection of the epipolar plane with the image plane.

epipolar plane

The plane, in space, containing a ground point and both exposure stations.

exposure station

During image acquisition, each point in the flight path at which the camera exposes the film. The exposure station has elements that define its position and rotation: X, Y, Z, omega, phi, and kappa.

exterior orientation

External sensor model information that describes the exact position and orientation of each image as they existed when the imagery was collected. The image’s position is defined as having 3D coordinates, and the orientation is defined as having three rotations that include omega, phi, and kappa.

exterior orientation parameters

The perspective center’s ground coordinates in a specified map projection and three rotation angles around the coordinate axes.

The process of identifying, delineating, and labelof natural and human-made phenomena from remimages.

feature draping

A condition that the feature follows the subtle chaof the terrain surface in the height dimension.

feature extraction

The process of studying and locating areas and oground and deriving useful information from ima

fiducial mark

Four or eight reference markers fixed on the frammetric camera and visible in each exposure. Fiducompute the transformation from pixel coordinatcoordinates.

Fixed Cursor Mode

A mode in which the 3D Floating Cursor remainposition while the image pair can be repositioned

Fixed Image Mode

A mode in which the image pair remains at a conwhile the 3D Floating Cursor moves freely in the

flight line

One of, typically, consecutive lines flown by an aiof exposure stations. A strip of photographs is caflight line.

flight path

The path of an airplane including, typically, multwith multiple exposure stations where the cameraPhotographs from several flight paths can be comblock of photographs.

239

focal length geolink

attribute data and

ip between the ) and the earth’s

satellites in cially designed th, can measure

ing satellites. The he geometric ate geodetic data da 2001).

ground wn.

p projection. pressed in feet or

being coordinate

GLOSSARY

The distance between the optical center of the lens and where the optical axis intersects the image plane. Focal length of a camera is determined in a laboratory environment.

focal plane

The plane of the film or scanner used in obtaining an aerial photo.

footprint

An outline corresponding to an image or image pair. Used to visualize areas of overlap between image pairs.

GCP

See ground control point.

geocentric

A coordinate system with its origin at the center of the earth ellipsoid. The Z-axis equals the rotational axis of the earth, the X-axis passes through the Greenwich meridian, and the Y-axis is perpendicular to both the Z-axis and the X-axis so as to create a 3D coordinate system that follows the right-hand rule.

geocorrect

A method of establishing a geometric relationship between imagery and the ground. Geocorrection does not use many GCPs, and is therefore not as accurate as orthocorrection or orthorectification. See also orthorectification.

geographic information system

(GIS) A unique system designed for a particular application that stores, enhances, combines, and analyzes layers of geographic data to produce interpretable information. A GIS may include computer images, hardcopy maps, statistical data, and any other data needed for a study, as well as computer software and human knowledge. GISs are used for solving complex geographic planning and management problems. A GIS consists of spatial data stored in a relational database with associated ancillary information.

A method of establishing a relationship between the features to which they pertain.

GIS-ready image

Imagery that has information about the relationshimage (as it existed when the image was recordedsurface.

global positioning system

(GPS) “A surveying method that uses a set of 24geostationary position high above the Earth. SpeGPS receivers, when positioned at a point on Earthe distance from that point to three or more orbitcoordinates of the point are determined through tcalculations of triangulation. GPS provides accurfor any point on Earth” (Natural Resources Cana

GPS

See global positioning system.

ground control point

(GCP) An easily identifiable point for which the coordinates of the map coordinate system are kno

ground coordinate system

A 3D coordinate system that utilizes a known maGround coordinates (X, Y, and Z) are usually exmeters.

ground point

Another term for the 3D Floating Cursor.

ground space

Events and variables associated with the objects photographed or imaged, including the referencesystem.

LYST FOR ARCGIS

image INS

as the focal al mark

and Remote

ment of photogrammetry 2000). For more .org>.

try and

ound the Z-axis.

re computed for a upon a set of s of probability unique values of y based on a large t kinds and

e time of data racy of the image

ht line without

USING STEREO ANA240

A picture or representation of an object or scene on paper or a display screen. Remotely sensed images are digital representations of the earth.

image center

The center of an aerial photo or satellite scene.

image pair

Two overlapping oriented images. A set of two remotely-sensed images that overlap, providing a 3D view of the terrain in the overlap area.

image scale

(SI) Expresses the ratio between a distance in the image and the same distance on the ground.

image space

Events and variables associated with the camera or sensor as it acquired the images. The area between perspective center and the image.

image space coordinate system

A coordinate system composed of the image coordinate system with the addition of a Z axis defined along the focal axis.

image-to-earth association

The 3D mathematical relationship between an image and the earth’s surface.

inertial navigation system

(INS) A technique that provides initial approximations to exterior orientation. This data is provided by a device or instrument. The instrument collects data about the attitude of the airplane in which it is located. The information it collects includes pitch (tilting forward and backward), roll (tilting sideways), and heading (the direction of flight) (National Oceanic and Atmospheric Administration 2001). See also omega, phi, kappa.

See inertial navigation system.

interior orientation

Describes the internal geometry of a camera suchlength, principal point, lens distortion, and fiducicoordinates for aerial photographs.

International Society of Photogrammetry Sensing

(ISPRS) An organization “devoted to the developinternational cooperation for the advancement ofand remote sensing and their application” (ISPRSinformation, visit the Web site <http://www.isprs

ISPRS

See International Society of PhotogrammeRemote Sensing.

kappa

In a rotation system, kappa is positive rotation ar

least squares adjustment

A technique by which the most probable values ameasured or indirectly determined quantity basedobservations. It is based on the mathematical lawand provides a systematic method for computingcoordinates and other elements in photogrammetrnumber of redundance measurements of differenweights.

lens distortion

Caused by the instability of the camera lens at thcapture. Lens distortion makes the positional accupoints less reliable.

line of sight

(LOS) The area that can be viewed along a straigobstructions.

241

LOS mosaicking

ide, to create a

r’s detectors.

positions above d for planimetric

f a point that has D data, Z.

graphs or images d. Neither the nternal geometry during image rements from a nts are in pixels

ogonally defined

eliberately offset

e X-axis.

GLOSSARY

See line of sight.

map coordinate system

A map coordinate system that expresses location on the earth’s surface using a particular map projection such as Universal Transverse Mercator, State Plane, or Polyconic.

mass points

Points whose 3D coordinates are known (X, Y, and Z), and which are used in creating a DEM or DTM. See also digital elevation model and digital terrain model.

metadata

A more highly organized or comprehensive level of data. Metadata files often contain lists of data files and auxiliary files such as header files, attribute files, and transform files. Metadata may be thought of as data about data.

metric photogrammetry

The process of measuring information from photography and satellite imagery.

mono

A view in which there is only one image. There are not two images to create an image pair. You cannot see in 3D using a mono view.

monocular vision

Vision using one eye.

monotonic

“Having the property either of never increasing or of never decreasing as the values of the independent variable or the subscripts of the terms increase” (Merriam-Webster OnLine Dictionary 2003).

The process of piecing together images, side by slarger image.

nadir

The area on the ground directly beneath a scanne

near vertical aerial photographs

Photographs taken from vertical or near vertical the earth captured by aircraft. Photographs are usemapping projects.

node

An element of a line or polygon, or the element ospecific coordinates in X, Y, and in the case of 3

nonoriented image pair

An image pair made up of two overlapping photothat have not been photogrammetrically processeinterior nor the exterior orientation, defining the iof the camera or the sensor as well as its positioncapture, has been defined. You can collect measunonoriented image pair; however, the measuremeand 2D.

nonorthogonality

The deviation from perpendicularity between orthaxes.

oblique photographs

Photographs captured by an aircraft or satellite dat an angle.

omega

In a rotation system, omega is rotation around th

LYST FOR ARCGIS

omega, phi, kappa orthocalibration

lacement and can nlike

ansformation age file (data)

and sensor ng from earth ion.

nt within acement, lens imagery is rthoresampling.

verlap, they share raphs taken along by 60 percent. ee also sidelap.

s seen in an aerial r coordinate ude or point of

nt in the image ates of the

. (3) After stem that defines

-axis.

USING STEREO ANA242

A rotation system that defines the orientation of a camera/sensor as it acquired an image. Omega, phi, kappa is used most commonly, where omega is positive rotation around the X-axis, phi is a positive rotation around the Y-axis, and kappa is a positive rotation around the Z-axis. This rotation system follows the right-hand rule.

optical axis

“The line joining the centers of curvature of the spherical surfaces of the lens” (Wolf and Dewitt 2000).

orientation

The position of the camera or satellite as it captured the image. Usually represented by six coordinates: X, Y, Z, omega, phi, and kappa.

ORIENTATION MANAGEMENT

(ORIMA) Software designed to process and produce data detailing orientation and triangulation in addition to bundle adjustment, etc. Output files can be imported into Stereo Analyst for ArcGIS.

oriented image

A first generation data product derived from imagery with a sensor model and spatial reference. Combining multiple oriented images allows for the creation of DTMs and collection of 3D features.

oriented image pair

An image pair with known interior (camera or sensor internal geometry) and exterior (camera or sensor position and orientation) orientation. The Y-parallax of an oriented image pair has been improved. Additionally, an oriented image pair has geometric and geographic information concerning the earth’s surface and a ground coordinate system. Features and measurements taken from an oriented image pair have X, Y, and Z coordinates.

ORIMA

See ORIENTATION MANAGEMENT.

A form of calibration that corrects for terrain dispbe used if a DEM of the study area is available. Uorthorectification, this method depends upon a trmatrix to resample on the fly thus leaving the imunaffected.

orthocorrection

A form of geometric correction that uses a DEMposition information to correct distortions resulticurvature and the like. See also orthorectificat

orthorectification

The process of lessening geometric errors inherephotography and imagery caused by terrain displdistortion, and the like. Then, the photography orresampled to a specified resolution. Also called o

overlap

In a traditional frame camera, when two images oa common area. For example, in a strip of photogthe flight path, adjacent images typically overlapThis measurement is sometimes called endlap. S

parallax

“The apparent angular displacement of an object aphotograph with respect to a point of reference osystem. Parallax is caused by a difference in altitobservation” (Natural Resources Canada 2001).

perspective center

(1) The optical center of a camera lens. (2) A poicoordinate system defined by the x and y coordinprincipal point and the focal length of the sensortriangulation, a point in the ground coordinate sythe sensor’s position relative to the ground.

phi

In a rotation system, phi is rotation around the Y

243

photogrammetric scanners principal point

pective center is

e position where rincipal point of 1983).

ipal point of “The point about ).

equired to restore ttings, and

earth is

nd scanning is er.

power of two and be displayed

m the principal on.

er, that detail the .

GLOSSARY

Special devices capable of high image quality and excellent positional accuracy. Use of this type of scanner results in geometric accuracy similar to traditional analog and analytical photogrammetric instruments.

photogrammetry

The “art, science and technology of obtaining reliable information about physical objects and the environment through the process of recording, measuring, and interpreting photographic images and patterns of electromagnetic radiant imagery and other phenomena” (American Society of Photogrammetry 1980).

pixel

Abbreviated from picture element. The smallest part of a picture (image).

planar

A constant elevation value (Z) is applied to each vertex of a feature.

plane table photogrammetry

Prior to the invention of the airplane, photographs taken on the ground were used to extract the geometric relationships between objects.

point

(1) A feature that has X, Y, and (sometimes) Z coordinates. A point can represent a feature such as a telephone pole. You can also collect multiple points to create a DEM or TIN. (2) In the case of defining the size of the 3D Floating Cursor used in the Stereo Window, a point equals a pixel.

point spacing

The distance between points sampled in terrain interpolation.

polygon

A set of closed line segments defining an area, composed of multiple vertices. Polygons can be used to represent features such as buildings, and can contain elevation values.

The point in the image plane onto which the persprojected.

principal point of autocollimation

Part of the definition of principal point, the imagthe optical axis intersects the image plane. The pautocollimation is near the principal point (Wolf

principal point of symmetry

Part of the definition of principal point, the princsymmetry can best compensate for lens distortion.which [distortions] are symmetrical” (Wolf 1983

project file

A SOCET SET® file containing the information rthe current state of a work. All necessary files, sepreferences are stored in the project file.

projection

The manner in which the spherical surface of therepresented on a flat (2D) surface.

pushbroom

A scanner in which all scanning parts are fixed aaccomplished by the forward motion of the scann

pyramid layer

An image layer that is successively reduced by a resampled. Pyramid layers enable large images tofaster at any resolution.

radial lens distortion

Imaged points are distorted along radial lines fropoint. Also referred to as symmetric lens distorti

rational polynomial coefficients

Coefficients, generally supplied by the data providposition of a satellite at the time of image capture

LYST FOR ARCGIS

raw image root mean square error

lculated solution henomena, a servation and a ulated value is

red. The sum of observations and alue.

lation functional mage space te system.

r distortions) that on (second order

reen—measured curate the screen sentations are on

etermine internal

ital value, and ation about the

USING STEREO ANA244

An image that does not have any projection associated with it. Raw images serve as a record of features, relationships between features, processes, and information.

reference coordinate system

A system that defines the geometric characteristics associated with events occurring in object space.

reference plane

In a topocentric coordinate system, the tangential plane at the center of the image on the earth ellipsoid, on which the three perpendicular coordinate axes are defined.

regular block of photos

A rectangular block in which the number of photos in each strip is the same. This includes a single strip or a single image pair.

rendering

Drawing an image in a view at the scale indicated by the zoom in or zoom out factor.

resample

The process of extrapolating data file values for the pixels in a new grid when the image is rescaled or rotated.

right-hand rule

A convention in 3D coordinate systems (X, Y, Z) that determines the location of the positive Z-axis. If you place your right hand fingers on the positive X-axis and curl your fingers toward the positive Y-axis, the direction your thumb is pointing is the positive Z-axis direction.

RMSE

See root mean square error.

(RMSE) Used to measure how well a specific, cafits the original data. For each observation of a pvariation can be computed between the actual obcalculated value. (The method of obtaining a calcapplication-specific.) Each variation is then squathese squared values is divided by the number of then the square root is taken. This is the RMSE v

rotation matrix

A three-by-three matrix used in the aerial triangumodel. Determines the relationship between the icoordinate system and the ground space coordina

rubber sheeting

A 2D rectification technique (to correct nonlineainvolves the application of a nonlinear rectificatior higher).

screen dot pitch

Screen dot pitch is the size of the pixels on the schorizontally in X and vertically in Y. The more acdot pitch values are, the more accurate scale reprethe screen.

self-calibration

A technique used in bundle block adjustment to dsensor model information.

sensor

A device that gathers energy, converts it to a digpresents it in a form suitable for obtaining informenvironment.

245

sensor model spatial reference

can be specified.

arily for its Z, or ins X and Y

sed to derive

image formed by n and parallactic

htly different

red on different vertical and the

otograph to the ). Makes viewing Y-parallax. See

lar vision.

GLOSSARY

A model describing the 3D relationship between a sensor, raw image, and the earth’s surface. A sensor model is required in order to create an oriented image, an image pair, a stereo model, and for the collection of 3D features. A sensor model makes an image oriented and photogrammetrically aware.

SI

See image scale.

sidelap

In a block of photographs consisting of a number of parallel strips, the sidelap is measured between the strips and is usually 20–30 percent in traditional frame camera photos. Sidelap is measured perpendicular to flight path.

single frame orthorectification

Orthorectification of one image at a time using the space resection technique. A minimum of three GCPs is required for each image.

softcopy photogrammetry

See digital photogrammetry.

space forward intersection

A technique used to determine the ground coordinates X, Y, and Z of points that appear in the overlapping areas of two or more images based on known interior orientation and exterior orientation parameters.

space intersection

A technique used to determine the ground coordinates X, Y, and Z of points that appear in the overlapping areas of two images, based on the collinearity condition. See also collinearity condition.

space resection

A technique used to determine the exterior orientation parameters associated with one image or many images, based on the collinearity condition.

A coordinate or other means by which a location

spot height

In feature collection, a point feature collected primelevation, value. A spot height feature also contacoordinate values.

stereo

A mode of visualizing wherein two images are uelevation information.

stereo model

The overlapping portion of an image pair. A 3D the brain as a result of changes in depth perceptioangles. See also image pair.

stereo pair

A pair of images of the same area taken from sligangles. See also oriented image pair.

stereo scene

Composed of two images of the same area acquidays from different orbits—one taken east of theother taken west of the nadir.

stereoscopic parallax

“The change in position of an image from one phnext caused by the aircraft’s motion” (Wolf 1983of 3D data possible, composed of X-parallax andalso X-parallax, Y-parallax.

stereoscopic viewing

Vision using two eyes. Also referred to as binocu

LYST FOR ARCGIS

strip of images/photographs TIN

r of the image on inate axes are

The X-axis is Z-axis is vertical

atical ptured it, and the

elevation points .

ra or sensor aerial

nsional matrix, y 1). Vectors , buildings, and

hree axes: X, Y, s to the elevation one vertex (such a polyline or

USING STEREO ANA246

In traditional frame camera photography, consists of images captured along a flight line, normally with an overlap of 60 percent for stereo coverage. All photos in the strip are assumed to be taken at approximately the same flying height and with a constant distance between exposure stations. Camera tilt relative to the vertical is assumed to be minimal. See also cross-strips.

support file

A SOCET SET® file containing photogrammetric metadata associated with an image in a project file.

tangential lens distortion

Distortion that occurs at right angles to the radial lines from the principal point.

Terrain Following Mode

A mode in which the 3D Floating Cursor follows the elevation of the terrain displayed in the Stereo Window. This is accomplished either by using an external elevation source, such as a DEM, or image correlation techniques.

thinning tolerance

A measure that prevents duplicate points within a certain distance in terrain interpolation (such as 5 meters).

threshold

Threshold is used during image correlation as a measure of probability that a points is the same in both the left image and the right image of an image pair. A high threshold value increases the probability of a correct match, but may take longer to process. Setting a low threshold increases the probability of a false match.

tie point

A point. Its ground coordinates are not known, yet it can be recognized visually in the overlap or sidelap area between two images.

See triangulated irregular network.

topocentric coordinate system

A coordinate system that has its origin at the centethe earth ellipsoid. The three perpendicular coorddefined on a tangential plane at this center point.oriented eastward, the Y-axis northward, and theto the reference plane (up).

transformation

A series of coefficients describing the 3D mathemrelationship between an image, the sensor that caground it has recorded.

triangulated irregular network

(TIN) A specific representation of DTM in whichcan occur at irregular intervals forming triangles

triangulation

Process of establishing the geometry of the camerelative to objects on the earth’s surface. See alsotriangulation.

vector

A point, line, or polygon. A vector is a one-dimehaving either one row (1 by j) or one column (i btypically represent objects such as road networksgeographic features such as contour lines.

vertex

A component of a feature, typically made up of tand (sometimes) Z. The Z component correspondof the vertex. A feature can be composed of onlyas a point as in a TIN) or many vertices (such as polygon).

247

vertical exaggeration

GLOSSARY

The effect perceived when a DSM is created and viewed. Vertical exaggeration is also referred to as relief exaggeration, and is the evidence of height differences in the stereo model.

vertices

More than one vertex, each with X, Y, and (sometimes) Z components.

X-parallax

The difference in position of a common ground point appearing on two overlapping images, which is a function of elevation. X-parallax is measured horizontally.

Y-parallax

The difference in position of a common ground point appearing on two overlapping images, which is caused by differences in camera position and rotation between two images. Y-parallax is measured vertically.

Z

The vertical (height) component of a vertex, 3D Floating Cursor, or feature in a given coordinate system.

Z-axis

In the image space coordinate system, the Z-axis is the optical axis. The image space coordinate system directs the Z-axis toward the imaged object. In object space, the Z-axis is orthogonal to the X and Y axes and is directed out of the earth’s surface.

LYST FOR ARCGIS
USING STEREO ANA248

References References

d Remote

: 1876. New

tion rchange.”

Society.”

igital Image (1975): 993.

mmetry and United , China, May

m-w.com/>.

-w.com/>.

.m-w.com/>.

: Inertial m/ins.html>.

artesian

This appendix lists references mentioned in this book.

American Society of Photogrammetry. 1980. Photogrammetric Engineering anSensing, XLVI:10:1249.

Asher & Adams. 1976. Asher & Adams’ Pictorial Album of American IndustryYork: Rutledge Books.

Free On-Line Dictionary of Computing. “American Standard Code for InformaInterchange from FOLDOC: American Standard Code for Information Inte24 Oct. 1999 <http://foldoc.doc.ic.ac.uk/foldoc>.

International Society for Photogrammetry and Remote Sensing. “ISPRS—The 29 May 2000 <http://www.isprs.org/society.html>.

Keating, T. J., P. R. Wolf, and F. L. Scarpace. 1975. “An Improved Method of DCorrelation,” Photogrammetric Engineering and Remote Sensing 41, no. 8

Konecny, G. 1994. “New Trends in Technology, and their Application: PhotograRemote Sensing—From Analog to Digital.” Paper presented at Thirteenth Nations Regional Cartographic Conference for Asia and the Pacific, Beijing1994.

Merriam-Webster OnLine Dictionary. “algorithm.” 07 Feb. 2001 <http://www.

Merriam-Webster OnLine Dictionary. “ellipsoid.” 29 May 2000 <http://www.m

Merriam-Webster OnLine Dictionary. “monotonic.” 28 Feb. 2003 <http://www

National Oceanic and Atmospheric Administration. “Inertial Navigation SystemNavigation System (INS).” 29 Mar. 2001 <http://www.csc.noaa.gov/crs/tc

Natural Resources Canada. “Carto Corner - Glossary of Cartographic Terms: Ccoordinate system.” 13 Jul. 2001 <http://www.atlas.gc.ca/english/carto/cartglos.html#4>.

249

Natural Resources Canada. “Carto Corner - Glossary of Cartographic Terms: coordinate system.” 13 Jul. 2001 <http://www.atlas.gc.ca/english/carto/cartglos.html#4>.

LYST FOR ARCGIS

Natural Resources Canada. “Carto Corner - Glossary of Cartographic Terms: GPS, Global Positioning System.” 13 Jul. 2001

c.ca/english/

rsity of

raw-Hill, Inc.

USING STEREO ANA250

<http://www.atlas.gc.ca/english/carto/cartoglos.html#4>.

Natural Resources Canada. “Carto Corner - Glossary of Cartographic Terms: parallax.” 13 Jul. 2001 <http://www.atlas.gcarto/cartoglos.html#4>.

Wang, Z. 1990. Principles of photogrammetry (with Remote Sensing). Beijing, China: Press of Wuhan Technical UniveSurveying and Mapping, and Publishing House of Surveying and Mapping.

Wolf, Paul R. 1983. Elements of Photogrammetry. New York: McGraw-Hill, Inc.

Wolf, Paul R., and Bon A. Dewitt. 2000. Elements of Photogrammetry with Applications in GIS. 3rd ed. New York: McG

IndexIndex

Symbols*.blk file 79

3Ddefined 233feature

1938, 160

366

147

134134

36

33135149

ode 147

tool 13736

163

defined 233*.img file

defined 234*.mxd file 119*.prj file 82

defined 234format 82

*.rrd file 19*.sde file 77*.sup file 82

defined 234format 83

“a” keyboard shortcut 149“c” keyboard shortcut 149“i” keyboard shortcut 149“r” keyboard shortcut 149“s” keyboard shortcut 54, 143, 149“t” keyboard shortcut 55, 141, 149“x” keyboard shortcut 149“y” keyboard shortcut 150“z” keyboard shortcut 149

Numerics1-Pane View 26, 1082D

defined 233to 3D conversion 35, 93

advanced options 95feature draping 95invalid elevations 100planar features 98point spacing 97thinning 98

characteristics 88collecting

and attributing workflow 15

defined 233Floating Cursor 132

accuracy of 133adjusting color 1adjusting size 13Auto Toggle Mode changing 46custom commands decrease elevation defined 233different shapes 1how it works 132how to position 1increase elevation keyboard shortcuts line width 136Manually Toggle Mrecenter 135use with Hyperlink

Floating Cursor tab 1model

characteristics 88defined 233

Position Toolapplying 48

Snapapplying 163customizing 166keyboard shortcuts

251

workflow 101virtual 89

2-Pane View 25, 109use in 3D Floating Cursor accuracy 133

options 162workflow 164

Snap tab 162to 2D conversion 31

LYST FOR ARCGIS

to 2D exporter 103 Analytical photogrammetry defined 235

79

4

Window 115

92, 231

173

ort

em

vice)

CD)

USING STEREO ANA252

virtual 893D Floating Cursor

recenter 1273-Pane View 25, 109

AAccuracy

CE90 145, 146LE90 145, 146

Active view alignment rotation mode 169

Addextension 14image pair 18toolbars 15

Additional parameter (AP)defined 234

Advanced options2D to 3D conversion 95

Aerialphotographs

defined 234triangulation (AT) 71, 231

defined 234Affine transformation 223

defined 234Airborne GPS

defined 234Algorithm

defined 234American Standard Code for Information

Interchange (ASCII)defined 234

Anaglyph 24, 27, 108, 1223D Floating Cursor color with 136defined 234

Analog photogrammetrydefined 235

defined 235AP (additional parameter)

defined 234ArcMap

data viewcalculating threshold 120image pair display 119orienting 21orienting displays 119

Display tab 119ASCII (American Standard Code for

Information Interchange)defined 234

AT (aerial triangulation) 231defined 234

At centroid 99Attribute

defined 235table

defined 235Attribution

defined 235Auto

-correlationdefined 235

Roam Modeactivating 27

Toggle 3D Floating Cursor Mode 50, 147, 157

limitations 147Automated terrain following

defined 235Automatic recenter 127Average interpolated 99Axis-To-Ground setting 43

BBinocular vision 106

Blockfile 80

characteristics defined 235importing 79

footprintdefined 235

of photographs 21defined 235

regular 244triangulation

defined 235Brightness

adjusting 28application in Stereo

Bundleblock adjustment 1

defined 235defined 235

Button mapping 45, workflow 174

CCache size 163Calibration certificate/rep

defined 236Cartesian coordinate syst

defined 236CCD (charge-coupled de

defined 236CE90 145, 146

applying 55Centroid

defined 236Charge-coupled device (C

defined 236Coefficient

defined 236

253

Collinearity Cartesian 221 image matching

107

)

port 173 173

ow 160

24

27 114 137

el) 107

el)

163ntation

INDEX

condition 228, 230, 236defined 236

defined 236equation 227

Colorblock 55, 133, 145, 156green 146red 146

Commandscustomizing 134Leica Feature Editing 134Stereo Analyst 134

Configuringsystem mouse 43

Constant elevationwith Virtual 2D To 3D 89

Context menu 157Continuous

Roam Modeapplying 115

Zoom Modeapplying 40, 117

Contrastadjusting 28application in Stereo Window 115effect on correlation 140stretch

defined 236types of 125

Control pointdefined 236extension

defined 236Convert

2D to 3D 34, 93workflow 94

3D to 2D 31, 103Features to 2D dialog 103

Coordinate system 218

defined 236geocentric 220ground 220image 219image space 219pixel 218topocentric 220

Coplanarity condition 206defined 236

Correlation 133coefficient

defined 237defined 236elevation bias 142image contrast 140terrain slope 139threshold 139, 237

Cross(3D Floating Cursor) 137-strips 237

Customizing tools 134

DDatum

defined 237Default Zoom

applying 39DEM (digital elevation model)

defined 237with Virtual 2D To 3D 89

Device 43, 173mapping buttons 45properties 43settings

accessing 42Digital

elevation model (DEM)defined 237

defined 237orthophoto

defined 237photogrammetry

defined 237stereo model (DSM)

defined 237terrain model (DTM

defined 237Digitizing

defined 237devices 173

adding 173choosing COM mapping buttonssupported 43

outside Stereo WindDirection of flight 214

defined 237Display

entire image pair 1overlap region 124polygon outlines 1

Docking Stereo WindowDot (3D Floating Cursor)Draping 89DSM (digital stereo mod

defined 237DTM (digital terrain mod

defined 237extraction 192

EEdge

keyboard shortcut Elements of exterior orie

defined 237Elevation 100

LYST FOR ARCGIS

bias 141, 142 F2 keyboard shortcut 51 Image Mode 156

n workflow

9

nt)

220

GIS)

s 183tion system) 1951800

ation 181

USING STEREO ANA254

invaliddefault value 100keep original 100minimum value 100

sourcedefined 238ensuring accuracy 37for 2D to 3D conversion 36in IMAGINE OrthoBASE 76Raster surface 36Virtual 2D To 3D 90

Ellipsoid 220defined 238

Endlapdefined 238

Endpointkeyboard shortcut 163

Epipolarcorrection 124line 206

defined 238plane 206

defined 238resampling 205

Exporter3D to 2D 103

Exposure station 214defined 238

Exterior orientation 73, 226defined 238parameters 238

External elevation sourceDEM 89TIN 89with Terrain Following Mode 138

FF10 keyboard shortcut 172

F3 keyboard shortcut 50, 149F4 keyboard shortcut 149Feature

collection3D Snap options 1623D snapping

workflow 164defined 238digitizing devices 173Monotonic Mode 172

workflow 172point feature 56polygon feature 49polyline feature 54squaring

workflow 171Squaring options 167with elevation bias 142workflow 158

drapingdefined 238

editingadjust location 64adjust vertex 61complete 135

extractiondefined 238

to 3D Options dialog 95Fiducial

mark 223defined 238

First line rotation mode 168Fixed

Cursor Mode 156applying 25definition 238feature collection workflow

160

defined 238feature collectio

158Flight

line 214defined 238

path 214defined 238

Focallength 73, 223

defined 239plane

defined 239Footprint

changing color 11defined 239

GGCP (ground control poi

defined 239Geocentric

coordinate system defined 239

Geocorrect 183defined 239

Geographicimaging 188information system (

defined 239Geolink

defined 239Geoprocessing techniqueGIS (geographic informa

3D GIS applicationsapplication 74, building blocks 18defined 239extracting 3D inform

255

-ready image 18, 71 nonoriented 241 K

09

149

1534

127e 169

147

INDEX

defined 239Global positioning system (GPS)

defined 239GPS (global positioning system)

defined 239Graphics card

information 122, 125specs 125supported with Stereo Analyst for

ArcGIS 24Ground

control point (GCP) 73, 191defined 239

coordinate system 220defined 239

pointdefined 239

spacedefined 239

IImage

Analysisusing to create oriented images

78center

defined 240coordinate system 219correlation

options 139with Terrain Following Mode

138defined 240pair

changing 23defined 240display 124invert 110

Pair dropdown listapplying 23

Pair Selection Toolapplying 23

scale 214scale (SI)

defined 240space

coordinate systemdefined 240

defined 240space coordinate system 219-to-earth association 71

defined 240types 190, 212

IMAGINE OrthoBASE 80calibration 76using to create oriented images 76

ImportIMAGINE OrthoBASE *.blk 80

process 81SOCET SET® *.prj file 82

process 84Inertial navigation system (INS)

defined 240INS (inertial navigation system)

defined 240Intelligent image 71Interior orientation 73, 223

defined 240International Society of Photogrammetry

and Remote Sensing (ISPRS)defined 240

Invert stereo model 110ISPRS (International Society of

Photogrammetry and Remote Sensing)defined 240

Kappadefined 234, 24

Keyboard shortcuts 143D Floating Cursor 3D Snap 163

LLE90 145, 146

applying 55Least squares

adjustment 229defined 240

Left imageapplying tools to 1

Leica Feature Editing 1Lens distortion 225

defined 240Line of sight (LOS)

defined 240Linear contrast stretch Longest line rotation modLOS (line of sight)

defined 240

MManually Toggle

3D Floating Cursor Map coordinate system

defined 241Mass points

defined 241Maximum

interpolated 99threshold 120

Metadatadefined 241

Metric photogrammetrydefined 241

LYST FOR ARCGIS

Midpoint phi, kappa 222, 226 Orthorectify 193

9

214

2 scanners

0

218

lated 99olated 99olated 99

USING STEREO ANA256

keyboard shortcut 163Min/max contrast stretch 126Minimum

interpolated 99threshold 120

ModeAuto Toggle 3D Floating Cursor

157Fixed Cursor 156Fixed Image 156Terrain Following 156

Monodefined 241

Monocular vision 106defined 241

Monotonicdefined 241Mode 172

Mosaickingdefined 241

NNadir

defined 241Near vertical aerial photographs

defined 241Node

defined 241Nonoriented image pair

defined 241Nonorthogonality

defined 241

OOblique photographs

defined 241Omega

defined 234, 241

defined 242Open

Cross (3D Floating Cursor) 137Cross with Dot (3D Floating Cursor)

137X (3D Floating Cursor) 137X with Dot (3D Floating Cursor)

137Optical axis

defined 242Orientation

changing 119defined 242

ORIENTATION MANAGEMENT (ORIMA)

defined 242Oriented

image 72defined 242process 70using Image Analysis for ArcGIS

78using IMAGINE OrthoBASE

76using SOCET SET® 82

image pairdefined 242

Orienting ArcMap display 21, 49ORIMA (ORIENTATION

MANAGEMENT)defined 242

Orthocalibration 76defined 242

Orthocorrectiondefined 242

Orthorectification 184defined 242single frame 229, 245

Overlapchanging color 11defined 242display only 124percentage 120,

Overlapping imagesthreshold 120

PParallax

defined 242Perspective center 219

defined 242Phi

defined 234, 24Photogrammetric quality

defined 243Photogrammetry 107

defined 243digital 237history 210metric 241plane table 243types of 188, 21uses 212

Photograph types 212Photography

terrestrial 221Pixel

coordinate system defined 243

Planardefined 243feature 89, 98

At centroid 99Average interpoMaximum interpMinimum interp

257

Plane table photogrammetry Recenter S

6

ation 229

43

3

230

INDEX

defined 243Point

defined 243spacing 97

defined 243Polygon

defined 243outlines 127

Principal point 219, 223defined 243of autocollimation

defined 243of symmetry

defined 243Project file 82

defined 243Projection

defined 243Pushbroom

defined 243Pyramid layer

creating 19defined 243

QQuad-buffered stereo 24, 27, 108,

1223D Floating Cursor color with 136

RRadial lens distortion 225

defined 243Rational polynomial coefficient 73

defined 243Raw

image 70defined 244

photography 182

3D Floating Cursor 127, 160Reference

coordinate systemdefined 244

planedefined 244

Regular block of photosdefined 244

Renderingdefined 244

Resample 76defined 244

Resolutionscanners 216

Right-hand rule 220

defined 244image

applying tools to 115RMSE (root mean square error)

defined 244Roam Mode

applying 39Root mean square error (RMSE)

defined 244Rotating stereo model 204Rotation 205

matrix 227defined 244

mode 168Active view alignment 169First line 168Longest line 169Weighted mean 168

RPC (rational polynomial coefficient) 73

Rubber sheetingdefined 244

Scalechanging 47

Scaling 205stereo model 204

Scanners 216Scanning resolution 21Screen dot pitch 125

defined 244SDE 77Self-calibration

defined 244Sensor

defined 244model 72, 191

defined 245SI (image scale) 214

defined 240Sidelap

defined 245percentage 214

Single frame orthorectificdefined 245

Slope 139Snap To Ground 143

applying 54use on buildings 1use on terrain 143

Snapping tolerance 16SOCET SET®

importing 82Space

forward intersection defined 245

intersectiondefined 245

resection 229defined 245

Spatial reference 72

LYST FOR ARCGIS

defined 245 visualization 107 options 14143 143

relation 139tion bias 141

97ce 98

ar network)

D 89

167

1 16, 113112

ystem 220

204

twork (TIN)

contrast stretch

USING STEREO ANA258

Spot heightdefined 245

Squaringoptions 167polyline 170rotation mode 168tab 167tolerance 167

Standard deviation of unit weightdefined 244

StereoAnalyst toolbar 16, 111defined 245display

contrast stretch 125linear 127min/max 126none 127two standard deviations

126displaying polygon outlines

127epipolar correction 124recentering the stereo cursor

127screen dot pitch 125

Display tab 124Enhancement toolbar 16, 113model 202

defined 245rotating 204scaling 204translating 204

pairdefined 245

scenedefined 245

View toolbar 16, 112

Windowapplying tools in

workflow 115context menu 157opening

workflow 24, 114using with data view 119views 24

1-Pane View 1082-Pane View 1093-Pane View 109

Stereoscopicparallax 202

defined 245viewing 106, 200

defined 245how it works 200

Stripof images

defined 246of photographs

defined 246Support file 82

defined 246Synchronize Geographic Displays

applying 49

TTangential lens distortion 225

defined 246Terrain

Following Cursor tab 53, 141Following Mode 156, 193

accuracy 146applying 53, 55, 144continuous 141defined 246methods of operation 138

preferences 1Snap To Groundusing image corusing with eleva

interpolationpoint spacing thinning toleran

slope 139Theta 224Thinning tolerance 98

defined 246Threshold 120

defined 246Tie point 192

defined 246TIN (triangulated irregul

defined 246with Virtual 2D To 3

Toleranceuse with Squaring

Toolbaradding 15Stereo Analyst 11Stereo EnhancementStereo View 16, StereoAnalyst 16

Topocentric coordinate sdefined 246

Transformation 73defined 246

Translating stereo modelTranslation 205Transparency 182Triangulated irregular ne

defined 246Triangulation

defined 246Two standard deviations

259

126 defined 247

INDEX

VVector

defined 246Vertex

defined 246keyboard shortcut 163

Vertical exaggeration 201defined 247

Verticesdefined 247

Views1-Pane View 1082-Pane View 1093-Pane View 109Stereo Window 108

Virtual 2D To 3D 89applying 92options 90process 90

WWeighted mean rotation mode 168

XX

(3D Floating Cursor) 137-parallax 202

defined 247screen dot pitch 125snapping tolerance 163thumb wheel

applying 145

YY

-parallax 203adjusting 142

screen dot pitch 125snapping tolerance 163thumb wheel

applying 145

ZZ

-axisdefined 247

Axis-To-Ground setting 43defined 247scroll wheel control 43, 44,

133, 147snapping tolerance 163thumb wheel

applying 25values

updating 93Zoom

adjusting 26In/Out By 2

applying 26To Data Extent 122

LYST FOR ARCGIS
USING STEREO ANA260

Documents

SAforArcGIS Web