LSGI521 L3 Spatial Data Acquisition in GIS BW.ppt · 9/14/2011 2 LSGI 521: Principles of GIS Lecture 3: Data Acquisition 1. Mount the map sheet on the digitizing table surface flat

9/14/2011

1

LSGI 521: Principles of GIS Lecture 3: Data AcquisitionLSGI 521: Principles of GIS Lecture 3: Data Acquisition

LSGI 521: Principles of GIS

Lecture 3: Spatial Data Acquisition in GIS

Dr. Bo Wu [email protected]

Department of Land Surveying & Geo-InformaticsThe Hong Kong Polytechnic University

LSGI 521: Principles of GIS Lecture 3: Data Acquisition

1. Learning outcomes2. Data acquisition: Manual digitization 3. Data acquisition: Field survey4. Data acquisition: Remote sensing and

photogrammetry

Contents

photogrammetry5. Data acquisition: LiDAR6. Data acquisition: GPS and other

9/14/2011 2


• By the end of this lecture you should be able to:– Describe how paper maps are digitized– Know the basic ideas about remotely sensing– Know the basic image process methods– Explain what is and how to process LiDAR data into

Learning Outcomes

GIS– Know what is GPS and its integration with GIS

9/14/2011 3


Data Input Stream in GIS

9/14/2011 4


• Digitizer – the equipment for digitizing• Operational procedure of digitization• Coordinate transformation• Error in digitization

Data Capture by Manual Digitization

g• Editing operation in digitization • Factors affecting digitization accuracy

9/14/2011 5


• Digitizing Table • Mouse • Paper map

Elements of a Digitizer

A small size digitizer

9/14/2011 6

A large size digitizer

Mouse with buttons and cross-hair

9/14/2011

2


1. Mount the map sheet on the digitizing table surface flat and secure

2. Identify points with known coordinates.– Tic marks or control points– Minimum two points depending on the software

P i t h ld b ll di t ib t d d d th ti t

Operational Procedure of Manual Digitization

– Points should be well distributed and covered the entire area to be digitized (usually, four corners of the map sheet)

3. Digitize control points and enter the true coordinatesinto the system

4. Software will compute coordinate transformationparameters and the residual– If residual is larger than the specified tolerance, you are

required to redo the entire procedure.

9/14/2011 7


5. Digitize Check Point– Check point is necessary to ensure that the map sheet

mounted on the surface has not been changed or drafted.

Operational Procedure of Manual Digitization (cont’d)

6. Digitizationo Two modes : point and

stream (time or distance)

9/14/2011 8

o Points digitized will be transformed from the digitizer local coordinates system to the known coordinates system using the computed transformation parameters


Map Digitization

9/14/2011 9

Digital map captured from digitizationMap before digitization


Coordinate Transformation Models

• Affine Transformation– Parallelity is still maintained, however the azimuth can be changed

y

x

• Similarity Transformation– Parallelity & azimuth of lines are maintained– 4 parameters

X' = a1x + a2y + a3

Y' = -a2x + a1y + a4

9/14/2011 10

• Projective Transformation– Parallelity and azimuth are not maintained– 8 parameters

X' =

Y' = 1 +y a + xaa +y a + xa

54

876

1 +y a + xaa +y a + xa

54

321

y

x

y

x

– 6 parameterX' = a1x + a2y + a3

Y' = a4x + a5y + a6


• Overshoots• Undershoots• Omitted edges or nodes• Duplicated edges or nodes• Mislocated edges or nodes

Errors in Manual Digitization

Duplicate edges

9/14/2011 11

+

+

+

+

Omitted edge

Duplicate Label

Overshooting

Duplicate edges

Undershooting


Possible Errors in Digital Map

Geometric Errors– A node is missing or

misplaced.– An edge is missing or

misplaced.

Topological Errors– Unconnected edges exist.– A polygon has a gap

between two edges, that is it is not closed.

9/14/2011 12

p– An edge has a bad shape

or too many (too few) points.

– A node has more than one position.

– Duplicate arcs are present.– A polygon has more than

one or no reference point associated with it (label point).

– A polygon may be missing.

9/14/2011

3


Factors Affect the Accuracy of Manual Digitization

1. Quality of original documente.g. drafting method, drafting media,

draftsman skill

2. Proper procedure of digitization and softwaree.g. digitization of control points,

9/14/2011 13

accuracy of control points,

digitization modes

3. Quality and resolution of the digitizere.g. resolution, repeatability, selectivity,

consistency

4. Operator's skill and concentration


Field Survey

• SurveyingHorizontal

– Traverse– Triangulation– Trilateration

Vertical

β

North

α

– Leveling

9/14/2011 16

A

B


Survey Instruments

Level and Staff

9/14/2011 17

Reflector Total StationElectronic Theodolite


• The collection of data without being in contact with the object.• There are usually two kinds of data we want to collect: thematic and

positional.• Thematic data tells us what the object is while positional data tells

us where the object is located.

Remote Sensing

9/14/2011 18


Energy Source & Spectral Bands

• Visible Bands– Blue (0.45-0.52 μm)– Green (0.52-0.60 μm)– Red (0.63-0.69 μm) – Panchromatic (0.50-0.90 μm)

9/14/2011 19


Spatial Resolution

9/14/2011 22

Ground sample distance (GSD) : The ground distance represented by the width of a pixel

9/14/2011

4


• Landsat TM (1982): 30 m• SPOT-5 (2002) : 5 m• IKONOS (1999):

– Panchromatic : 1 m– Multispectral : 4 m

• QuickBird (2001):

Spatial Resolution of Typical Sensors

China’s first survey satellite willQuickBird (2001):– Panchromatic : 0.61 m– Multispectral : 2.4 m

• GeoEye (2008): – Panchromatic : 0.41 m– Multispectral : 1.65 m

9/14/2011 23

China s first survey satellite will be launched in 2011 with the resolution of 2.5 m and 4 mChina’s Chang’E-1 CCD Camera (2007): 120 mNASA’s Lunar Reconnaissance Orbiter Camera (2009): 0.5 mNASA’s Mars Reconnaissance Orbiter Camera (2006): 0.3 m


Spectral Resolution

The narrower the wavelength bands the sensor can detect, the higher the sensor’s spectral resolution

• Panchromatic sensors– To record the overall brightness of a scene– A camera with black and white film

• Multispectral sensorsp– Simultaneously record multiple wavelength bands– Landsat Thematic Mapper: seven multispectral bands– IKONOS and QuickBird: four multispectral bands

• Hyperspectral– Simultaneously record data in hundreds of contiguous, very narrow

spectral bands in UV, visible, NIR, mid-IR bands– EO-1: 254 bands, 10 nm resolution

9/14/2011 24


Panchromatic Image

Tampa Bay, FL

9/14/2011 25


Multispectral Images

9/14/2011 26


Multispectral Image Analysis

9/14/2011 27


Hyperspectral Imagery

9/14/2011 28

Band 1, 399 nm Band 100, 1310 nm Band 206, 2466 nm

… … … …

9/14/2011

5


• Airborne

– Aerial photographs

– SAR, InSAR

• Satellite

Remote Sensing Platform

• Satellite

– Landsat, SPOT, IKONOS, QuickBird

• Ship/boat

– Sonar

– Underwater Video

9/14/2011 29


Airborne Remote Sensing

The first form of remote sensingEquipment :

• Cameras, • Photographic films, and • Camera filters.

Nevertheless recently high electronic system, the aerial photograph will ti t b i f f t i

9/14/2011 30

continue to be primary form of remote sensing.


• Use Fixed-wing aircraft• Flying over the area in the zigzag pattern• Flight attitude : 3,000~4,000m• Scale : 1:20,000~1:24,000• Sidelap : 30%• Overlap : 60%

Airborne Remote Sensing

Overlap : 60%

All ground feature will appear on at least two photo.

31Pushbroom CCD Camera

9/14/2011


Stereo Image Processing & 3D Reconstruction

S1

a2

a1

A

B

b2

b1

9/14/2011 32

S2


Aerial Photogrammetry

g61 g62

9/14/2011 33


• Interior Orientation• Exterior Orientation• Aerial Triangulation• Collinearity Equations

)()()( ZZRYYRXXR −+−+−

Aerial Photograph Geometry

)()()()()()()()()()()()(

333231

2322210

333231

1312110

sss

sss

sss

sss

ZZRYYRXXRZZRYYRXXRfyyy

ZZRYYRXXRZZRYYRXXRfxxx

−+−+−−+−+−

−=Δ−−

−+−+−−+−+−

−=Δ−−

⎥⎥⎥

⎦

⎤

⎢⎢⎢

⎣

⎡

−++−−+−+

=ϕωϕωϕ

κωκϕωκωκϕωκϕκωκϕωκωκϕωκϕ

coscoscossinsincossinsinsincoscoscossinsinsinsincossinsincossincossincoscossinsincoscos

R

))((2))(2())((

))((2))(2())((

0012

02

24

22

10

0022

02

14

22

10

yyxxpyyrprkrkyyy

yyxxpxxrprkrkxxx

−−+−+++−=Δ

−−+−+++−=Δ

349/14/2011

9/14/2011

6


• Most satellite sensors detect electromagnetic radiation (EMR) electronically as a continuous stream of digital data. The data are transmitted to ground reception stations, processed to create defined data products, and made available for sale to users on a variety of digital data media.

Satellite Remote Sensing

9/14/2011 35


Application of Satellite Images

Deforestation in Bolivia

1975 1992 2000

1973

Mississippi Delta sediment deposition

1989 20039/14/2011 36


High-Resolution Satellite Imagery

IKONOS Image 1 m Resolution

QuickBird Image0.6 m Resolution

Mars HiRISE Image

0.3 m Resolution

9/14/2011 37


Satellite Image Geometry

Rational Function Model

9/14/2011 38


• Image interpretation – The detection, identification, and measurement of specific

features

• Digital image classification– Classification of the total scene content into a limited

number of major classes

Deriving Information from Remote Sensing Images

number of major classes

• Data transformations– Mathematical operations on satellite image spectral data

that enhance the interpretation or discrimination of selected terrain features

• Change detection– Detection of numerical differences in corresponding pixel

values between dates 9/14/2011 39


• LiDAR• RADAR• SAR• SONAR

Other Remote Sensing Technologies

Questions:Do you know about any of these technologies?

What are these abbreviations stand for? Difference among them?

9/14/2011 40

9/14/2011

7


Data Capture by Laser Scanning

• Terrestrial scanning

• Airborne scanning

9/14/2011 41


LiDAR

• LiDAR = Light Detection and Ranging• Pulse of LASER light emitted• Return time and intensity measured• Converted into 3D coordinates

9/14/2011 42


LiDAR Data Format (LAS)

9/14/2011 43

ALS Points Projected on the Orthophoto


9/14/2011 44


Cyber City Modelling from LiDAR Data

9/14/2011 45

Scanned Urban Area

3D Cyber City Model


Object Detection from LiDAR Data

9/14/2011 46

3D visualisation of the LiDAR points Photo taken from the site

9/14/2011

8


RADAR and SAR

RADAR = Radio Detection and Ranging A RADAR system has a transmitter that emits either microwaves or radio waves that are reflected by the target and detected by a receiver. Although the signal returned is usually very weak, the signal can be amplified. This enables RADAR to detect objects at

h h i i h

SAR = Synthetic Aperture Radar SAR is a form of radar in which the large, highly-directional rotating antenna used by conventional radar is replaced with many low-directivity small stationary antennas scattered over some area near or around the target area. The many echo

f i d h diff

9/14/2011 47

ranges where other emissions, such as sound or visible light, would be too weak to detect.

waveforms received at the different antenna positions are post-processed to resolve the target.


• SONAR = Sound Navigation And Ranging• Sound waves travel in water at about 1,500

m/sec.• Types

– Side-scan sonar

SONAR

Side scan sonar– Single-beam sonar– Multi-beam sonar

9/14/2011 48


• Global Positioning System (GPS) – American version of satellite positioning system– Initiated in 1973– 24 satellites.

• Global Navigation Satellite System (GLONASS)– The Russian version of satellite positioning system– Initiated in 1982

Global Positioning System (GPS)

– 17 satellites

• Galileo Positioning System– Europe version of satellite positioning system– Initiated in 2005– Expect 30 satellites (1 on orbit now)

• BeiDou Positioning System– Chinese version of satellite positioning system– Initiated in 2000– Expect 35 satellites (4 on orbit now)

499/14/2011


GPS Instruments

About US$300 in 1999

9/14/2011 50

About US$3,000 in 1999


• Mathematically, we need four satellite ranges to determine exact position.

How GPS Works

529/14/2011


Position Measurement

1 2

53

3 • The forth measurement from the forth satellite is necessary to determine the final position of the object

• But usually three ranges are enough because one of the two points is a ridiculous answer (either too far from Earth or moving at an impossible velocity) and can be rejected without a measurement

9/14/2011

9/14/2011

9


Use GPS for Positioning

9/14/2011 54


• Remote Sensing Tutorial: http://rst.gsfc.nasa.gov/Front/tofc.html

• Aerial Photography and Remote Sensing: http://www.utexas.edu/depts/grg/gcraft/notes/remote/remote.html

Further Readings

• Digital Image Processing: http://www.cla.sc.edu/geog/rslab/rsccnew/

• Earth Science Pictures: http://pao.gsfc.nasa.gov/gsfc/earth/pictures/earthpic.htm

• GPS (highly recommended): http://www.trimble.com/gps/index.htm

9/14/2011 55


• Summarization of the main ideas presented in this lecture:

Review

• Questions?

9/14/2011 56

Documents

LSGI521 L3 Spatial Data Acquisition in GIS BW.ppt · 9/14/2011 2 LSGI 521: Principles of GIS Lecture 3: Data Acquisition 1. Mount the map sheet on the digitizing table surface flat