BeeGIS

Development of Open Source tools for Digital Field

Mapping

User manual

Universita degli Studi di Urbino “Carlo Bo”Dottorato di ricerca - XXII ciclo

November 2009

AKNOWLEDGEMENTS

This work is dedicated to all the sparkling knights of the open source community, particularly inOsGeo. You all gave so much to me, got the best of me.

***A particular salute goes to the round table of the JGrass, uDig, BeeGIS, Geotools and GRASS

community. Can’t tell the nights and weekends we spent together hacking for good...***

The biggest thanks go to Silli, for being THE WonderSilli, for supporting me in this, fordemonstrating once more that what we are building with HydroloGIS is something really really

different...***

Kudos to my parents, for teaching me that the only really perfect investment is the one thatresults in knowledge, passion and fun (in scrambled order).

***Thanks to Mauro, for passing over his ideas and giving me space and advice in all this. Keep your

digital mapping dreams going, they do work!

Readme

This manual was written by Andrea Antonello as part of his PhD under the tutorship of prof.Mauro De Donatis. Various chapters of the original thesis were removed from the manual.It is distributed under CREATIVE COMMONS license:You are free:

• to Share - to copy, distribute and transmit the work

• to Remix - to adapt the work

Under the following conditions:

• Attribution - You must attribute the work in the manner specified by the author or licensor(but not in any way that suggests that they endorse you or your use of the work).

With the understanding that:Waiver - Any of the above conditions can be waived if you get permission from the copyrightholder.Other Rights - In no way are any of the following rights affected by the license:

• Your fair dealing or fair use rights;

• The author’s moral rights;

• Rights other persons may have either in the work itself or in how the work is used, such aspublicity or privacy rights.

Notice - The complete license description can be found at the following link:http://creativecommons.org/licenses/by/3.0/legalcode

6

Contents

1 Introduction 1

1.1 Digital field mapping . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2

1.2 How the history of the survey can lower the uncertainty . . . . . . . . . . . . . . . . 3

1.3 The hardware . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4

3 BeeGIS 21

3.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21

3.2 The connection to the GPS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21

3.2.1 The GPS toolbar . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21

3.2.2 Setting up the gps connection . . . . . . . . . . . . . . . . . . . . . . . . . . . 23

3.2.3 Data logging . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 25

3.2.4 Gps points acquisition tools . . . . . . . . . . . . . . . . . . . . . . . . . . . . 26

3.2.5 Layer creation aid tools . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 27

3.3 The Geonotes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 28

3.3.1 Anatomy of a geonote . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 28

3.3.2 The Paint box . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 30

3.3.3 The Text box . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 31

3.3.4 The Media box . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 31

3.3.5 Geonotes tools . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 33

3.3.6 The fieldbook view . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 34

3.4 The Annotation Layer . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 42

3.5 The embedded database . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 44

3.5.1 H2 database . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 45

3.5.2 The embedded database view . . . . . . . . . . . . . . . . . . . . . . . . . . . 47

3.5.3 The database structure . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 49

3.6 Other tools . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 50

3.6.1 Gps data logs export . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 50

i

CONTENTS

3.6.2 Photo import . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 53

3.6.3 Geopaparazzi import . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 56

4 A lightweight solution for fast surveys 58

4.1 The Android platform . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 58

4.1.1 Hardware features . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 59

4.2 Geopaparazzi . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 60

4.2.1 Georeferenced notes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 62

4.2.2 Georeferenced pictures . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 64

4.2.3 Gps logging . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 67

4.2.4 The OpenStreetMap view . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 69

4.3 Export features . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 75

4.3.1 Kml Export for Google Earth . . . . . . . . . . . . . . . . . . . . . . . . . . . 75

4.3.2 The disk data structure . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 75

4.4 Upgrading to the desktop GIS, importing to BeeGIS . . . . . . . . . . . . . . . . . . 76

4.4.1 Notes import . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 81

4.4.2 Gps logs import . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 81

4.4.3 Pictures import . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 81

5 Use case: verification and update of datasets needed for the evaluation of the

environmental impact of the production of hydro-electricity 84

5.1 The objectives of the survey . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 84

5.2 Preparation of the cartography for the survey . . . . . . . . . . . . . . . . . . . . . . 84

5.2.1 Raster cartography . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 85

5.2.2 Vector cartography . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 85

5.3 Preparation of the tablet pc . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 86

5.4 Field mapping . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 87

5.4.1 Use of the GPS and geonotes . . . . . . . . . . . . . . . . . . . . . . . . . . . 87

5.5 Back in the office . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 89

ii

1 Introduction

This work convolves different needs and efforts into the creation of a set of tools dedicated to

digital field mapping.

From the begin the priority has been set on the need to be able to collect informations in

support to research in the environmental field. These informations need to be properly organized

and cataloged, in order to be used in future processes of evaluation and formation of scientific

theories. The importance of this matter is further discussed in section 1.1.

The approach of the digitalization of the field mapping has been accurately discussed in the last

years and is gathering a solid user base due to its already quite mature implementation in different

software tools. The primary factor for this push can be attributed to the fast growing hardware

market, that in the last years has been able to provide any average user and professional with

cheaper tools for digital mapping as for example tablet computers and gps receivers.

The availability of such hardware at a limited cost has accelerated research in that field and

pushed with force all those schools of thought that center their attention not only on the final

product of the mapping, i.e. the map itself, but also on the archiving as much correlated information

of the mapping itself, as possible. These doctrines teach that the final product is completely centric

to the surveyor in terms of prior scientific knowledge, experience, type of survey taken. In the case of

maps created at a national or regional scale, this involves several surveyors with different scientific

background, hence surveying techniques and interpretation of the environment. This raises the

need for a clear distinction between data and knowledge and of a way to store or archive certain

information bound to the implicit knowledge of the mapping expert, which otherwise get lost in

time and can’t be evinced from the final map product.

For this reason the design of a set of tools to support a surveyor in his digital field mapping has

been approached in this work. The implementation of the designed tools has been done inside the

framework of a geographic information system, which came as an obvious choice, given the fact

that the first goal is that to geographically localize all the collected information to assure spatial

(and temporal) coherence.

The created instruments, that will be thoroughly described in the following chapters, can be

summarized in a GIS framework, named BeeGIS (see chapter 3), for which extensions for digital

mapping were developed, in order to be used as a digital fieldbook and be able to connect to a gps

device for proper localization.

Depending on the kind of survey that has to be done, the hardware unit loading BeeGIS can

be different. For field mapping under heavy meteorologic conditions the hardware market provides

1. Introduction

rugged portable computers, that have particular protection systems for the connected devices.

Instead for the urban mapping of the electric lines network a normal tablet computer, has proved

to be extremely useful, since in that case the weight of the device and its screen reflectivity are

the features to look at. Some tests were done also with ultra mobile tablet pcs, which dimensions

are placed between the tablet pc and the PDA (personal digital assistant). While the screen of

the PDA is too small for most of the mapping tasks, the ultra mobile has proved to be a good

compromise between usability and ease of transport.

For those cases in which it is important to be able to quickly collect data without having the

possibility to rely on the tablet computer or when it is possible to use only one hand under heavy

conditions, a light version has been created, named Geopaparazzi (see chapter 4). The platform

to work on has been chosen to be smartphone devices based on Google’s open sourced Android

operating system. The phone has been chosen as the target device because of its size and mainly

because it is widespread and meant to be always at hand.

Part of the work done bases on passed experiences gathered by the team on the matter of digital

field mapping. In fact BeeGIS builds on the ashes of a project called MapIt (see [De Donatis et al., 2006]

and [De Donatis et al., 2005]) 1, which after a very promising start, was blocked due to incompre-

hension between academic and industrial environments and went into end of life.

This work objective is twofold, probably due to the fact that is contains both the views of an

environmental engineer and a geologist. It presents new tools for digital field mapping that on one

hand aim to solve some of the problems bound to the uncertainty implicit in path from the field

mapping to the final map product, on the other hand show the advantage to survey directly in

digital mode, which makes certain processes quicker and less error-prone.

1.1 Digital field mapping

Digital field mapping is the process of surveying using a portable computer unit connected to

a gps for the geolocalization. At the time of writing many tablet pcs already come with gps units

integrated. Up to now it is often still necessary to connect the gps unit via bluetooth or even usb

port, which makes it a bit less comfortable to carry around.

The aim of the mapping is the collection and persistence of as many spatial data as possible,

bound to the interpretation of the mapping expert. In fact being able to persist the interpretation,

the implicit knowledge of the expert is the key point to the modern digital mapping. Digital

mapping is often confused with simply installing a GIS on a tablet pc and taking it out in the field

1http://www.uniurb.it/ISDA/MAPIT/index.htm

2

1. Introduction

for a survey. This might end up in defining some layers containing point features to remember

places or layers of polygons to define a particular area of interest. This can all be taken to its limit

and be done in particular accurate way, but represents few percentage of the real power that is

behind digital field mapping.

It is extremely important to be able to collect not only the more standard data but also the

interpretation of the expert. This raises the need for tools to take pictures, draw sketches on

the pictures to persist impressions regarding the picture, draw sketches on the map, create media

documents and store them in geographic positions, add textual notes and even spoken notes and

whatever else comes to the surveyor’s mind.

The final objective is collect up to were possible all the data and knowledge that lead to the

creation of the final result: the map.

1.2 How the history of the survey can lower the un-

certainty

The key point of the digital field mapping is bound to the concept of persistence not only of the

measured data but also of the interpreted data.

The problem bases on the fact the paper based maps are not able to present their generation

history, only a fraction of the collected information is shown on the map. The main problem in

this is that there is often now way to gather the information about the reasoning used, which is

hidden behind the prior knowledge and experience of the mapping expert[Jones et al., 2004].

Another limiting factor of paper maps is the fact that they are bound to the scale at which they

were produced. A change of scale can produce incongruence and major errors. This is not the case

of digital maps, which are produced with an elevated accuracy. For example on a paper map at

a scale of 1:10000 the pencil stroke of 1 millimeter on the map represents 10 meters in the real

world, drawing shifted by a half centimeter introduces 50 meters of shift. Even worst in the case

of 1:50000 map where the errors grow to some hundreds of meters. The digital survey is bound to

the gps acquisition, which is nowadays below the 5 meters shift and mostly below the meter shift.

These examples explain the need for tools able to persist the history of the survey in terms

of spatiality and timespan. A set of GPS measures already describes a spatial and temporal

distribution, which already explains some of the the reasons for a articular way of interpreting the

environment. For example starting a survey from the top of a valley or from the bottom most

probably leads to very different results. This resides in the obvious fact that certain indicators are

seen before others and the interpretation evolves along that line. Spatial and temporal information

3

1. Introduction

is not enough, it is necessary to also persist all that knowledge that is needed to the surveyor itself

to understand the reasoning used to produce a map even after years, and even more important to

give the possibility to other surveyors to build on the hypotheses already elaborated when a map

has to be updated in the years. If information is lost in time, every time a particular product is

not understood, it is discarded and done from scratch.

1.3 The hardware

The current trend of all major pc producers toward tablet pcs and touch devices will enhance

the digital field mapping a lot in the next years. Already during the few years of this work hardware

has changed quite a lot, as for example in terms of usability, battery time and screen luminosity.

The hardware tested and used during this work is the following:

• Xplore iX104: The Xplore (see figure 1.1) is a rugged tablet that can be used under heavy

meteorological conditions and was tested to be working well also with rain. Xplore itself is

one of the most advanced brands for rugged tablet pcs which supplies the most innovative

technologies for outdoor mapping. It has been a real pity that the only model available

for testing purposes was the old iX104, which was about 6 years old (please remember that

6 years in todays constant and fast evolution of the mobile technologies represents ages).

Nevertheless it has been the tablet pc on which most of BeeGIS has been developed and has

perfectly served the purpose. Since the model was quite old, it came with no internal gps,

which instead had to be connected by usb or by usb bluetooth dongle. Indeed eventually

the usb port broke since the dongle was too exposed. Newer models all come with internal

bluetooth and anyways the Xplore has an own expansion GPS module that can be attached

to the tablet pc through safe rugged connections.

Main specifications:

– Intel Pentium III-M @ 866 MHz

– 512 MB DDR RAM

– 40GB HDD

– 10.4” display XGA Transmissive LCD

– weight: 2 Kg

– size (WxHxD): 283.9/209.3/40.8mm

– Windows XP Tablet PC edition

4

1. Introduction

Notes:

– to survey a whole day it is necessary to have two batteries

– the screen is not well usable on bright sunny days without covering it

– the Xplore doesn’t have a keyboard

Figure 1.1: The Xplore rugged tablet pc.

• Asus R2E: The Asus (see figure 1.2) is an ultramobile tablet pc. While based on a minor

processor, it supplies enough power to run the GIS on windows vista and be well usable. A

great advantage of the device is the fact that it has an integrated gps. The smaller screen

and the missing need for a bluetooth connection help letting the battery last longer. Another

5

1. Introduction

advantage of the pc is the possibility to insert mobile internet cards and be online during the

survey. This opens to the possibility to work on centralized datasets.

Specifications:

– Intel A110 (Stealey) @ 800 MHz

– 1024 MB RAM

– 80GB HDD

– 7” display CCFL b/l, Heavy (Stylus) Touch

– weight: 0.96 Kg

– size (WxHxD): 234/133/28 mm

– Windows Vista Home Premium

– integrated wireless, bluetooth and HSDPA

– integrated camera (1.3 megapixels)

Notes:

– the screen is almost unusable on bright sunny days without covering it

– the pc comes with an external portable keyboard

– the integrated camera can’t be used for taking pictures when on the field, since it is

thought as a webcam for internet video chat use and is placed on the front side of the

pc

• HP Compaq 2710p: Given the fact that during this work mostly engineering surveys were

done, the HP (see figure 1.3) has been the most satisfying device used. It is very light, has

a luminous screen and can be used even under bright sun conditions. It also can be used as

normal portable pc, since it has a keyboard and a good processor. Also the ram memory of

two gigabytes makes it very usable. This is the newest pc we were able to test and explains

also the trend. The new tablet pc is way more usable in terms of screen and battery than

the older rugged pc which was much more expensive than the HP tablet. This shows the gab

that few years filled in the hardware field.

Specifications:

– 1.2-GHz Intel Core 2 Duo Ultra Low Voltage U7600

– 2 GB RAM

– 80GB HDD

6

1. Introduction

Figure 1.2: The Asus ultra mobile computer.

7

1. Introduction

– 12.1” display in TFT active matrix

– weight: 1.7 Kg

– size (WxHxD): 289/210/28 mm

– Windows Vista Home Premium

– integrated wireless, bluetooth and HSDPA

– integrated sdcard reader

Figure 1.3: The HP Compaq 2710p tablet computer and the HTC Magic Android phone.

• HTC Dream: The HTC dream is the first smartphone on the market based on the open

source Android platform. Notes:

8

1. Introduction

– a battery with gps logging continuously enabled lasts for about 4 hours

– the phone has a slide out keyboard

Specifications:

– Qualcomm MSM7201A, 528 MHz

– 192 MB RAM

– extensible slot for microSD memory card

– 3.2-inch TFT-LCD flat touch-sensitive screen with 320 x 480 (HVGA) resolution

– weight: 158 gr

– size (WxHxD): 117.7/55.7/17.1 mm

– Android operating system

– integrated wireless, bluetooth, GPS, Digital Compass, Motion Sensor

– integrated 3.2 megapixel color camera with auto focus

– Slide-out 5-row QWERTY keyboard

• HTC Magic: The HTC Magic (see figure 1.3) is the successor of the Dream model. While

some enhancement in the RAM memory and the bluetooth were made, the processor didn’t

change. Notes:

– a battery with gps logging continuously enabled lasts for about 4 hours

– the phone has a slide out keyboard

Specifications:

– Qualcomm MSM7200A, 528 MHz

– 288 MB RAM

– extensible slot for microSD memory card

– 3.2-inch TFT-LCD flat touch-sensitive screen with 320 x 480 (HVGA) resolution

– weight: 116 gr

– size (WxHxD): 113/55.56/13.65 mm

– Android operating system

– integrated wireless, bluetooth, GPS, Digital Compass, Motion Sensor

– integrated 3.2 megapixel color camera with auto focus

9

2. BeeGIS

2 BeeGIS

2.1 Introduction

BeeGIS had one requirement: support digital field mapping. As such the development work to

be done was to create extensions for the uDig framework that would serve the tasks of digital field

mapping. As the next chapters will analyze and explain, several tools were created and integrated

in uDig, ranging from GPS acquisition support to tools for the collection of diverse informations

in the field, like for example the geonotes.

The next chapters will explain the developed tools and their usage.

2.2 The connection to the GPS

2.2.1 The GPS toolbar

The gps engine has a dedicated set of icons in the main toolbar, as can be seen in figure 3.1.

Figure 2.1: For every imported picture a geonote is created and the picture is stored inside the mediabox of thegeonote.

The single tools will be discussed in the following section, a short description of every tool is

hereby presented:

• GPS settings - opens the gps settings dialog

Figure 2.2: gps settings

• Toggle logging - toggles between activating and deactivating data acquisition

Figure 2.3: toggle gps logging

• Manually add point - adds a point to the selected layer from the current gps position

10

2. BeeGIS

Figure 2.4: manually add point

• Add geonote - adds a new Geonote placed at the current gps position

Figure 2.5: add geonote

• Automatic point acquisition - toggles automatic point acquisition from gps, adding the points

to the selected layer

Figure 2.6: automatic point acquisition

• Toggle center on gps - toggles the centering on the current gps position, whenever the gps

would get out of the map’s viewport

Figure 2.7: center on gps

• New point layer - creates a new point layer with default attributes

Figure 2.8: new point layer

• New line layer - creates a new line layer with default attributes

Figure 2.9: new line layer

• New polygon layer - creates a new polygon layer with default attributes.

Figure 2.10: new polygon layer

11

2. BeeGIS

2.2.2 Setting up the gps connection

The first thing to do to get started with gps data acquisition is to connect the gps to BeeGIS (this

assumes that the gps has been connected to the operating system first). Since an unexperienced

user might not know how to do that, several ways have been implemented in order to get to the

gps settings panel:

1. the gps settings icon in the gps toolbar

2. in the case the gps logging is started without having previously connected the gps, the gps

settings panel is proposed

3. in the main preferences, there is a gps entry that proposes the gps settings panel.

The gps settings panel can be divided in 5 different sections, as shown in figure 3.11:

1. the gps port section. To properly work the port to which the gps is connected has to be

configured. To help the user in this process the search button triggers a scan of the available

possible ports in the system. The available ports are then presented to the user in the panel

shown in figure 3.12.

2. once the port has been selected, the gps connection can be performed by pushing on the start

gps radio button.

3. if the gps connection properly occurred, the textarea in this section will show a sample NMEA

string to test if the data arriving from the port are really gps data (see figure 3.13).

4. in this section two parameters can be set. The first is the data acquisition interval in seconds,

which defines the time interval between the acquisition of a gps impulse and the next. This

is necessary, since the gps provides data in a continuous way. The second parameter defines

the minimum distance between two subsequent gps point, in order to not assume that the

point is the same. This is important in the case in which the user stops during the survey

and he doesn’t want to collect a lot of points in the same position. it is very important to

note that this value is given in the unit of the current coordinate reference system.

5. For development purposes a section has been added, to simulate a virtual gps without the

need of being connected to a real gps and being in open space to make the receiver work. By

activating that switch and starting the gps, the NMEA data will be taken from looping over

a sample file.

12

2. BeeGIS

Figure 2.11: The gps settings window, as it appears at the first start.

Figure 2.12: The ports search utility started from the gps settings window.

13

2. BeeGIS

Figure 2.13: Once the start button is pushed, the textarea displays an extract of what enters from the chosen port.

2.2.3 Data logging

Once the gps has been connected to BeeGIS, it is possible to start data logging from the toolbar

through the button of figure 3.3.

Once data logging is enabled, also all the gps toolbar buttons are enabled and can be used.

When enabling logging, the user will be prompted to decide whether to start a new feature or

continue from the last feature on the layer, in the case of further creation of new points, lines or

polygons on the selected layer. This will be asked every time logging is enabled.

Once logging is enabled, the gps view is opened, reporting the currently selected data from the

gps (see figure 3.14).

The logging itself doesn’t modify any data on the layers, it primarily visualizes the gps position

on the map view. The gps position symbol is as a cross which is traversed by an arrow, that

indicates the direction of movement of the gps (see figure 3.15).

Figure 2.15: The gps symbol visualized when logging is active.

If the gps position isn’t visualized, it is possible that the gps position is outside the map viewport.

This can be solved by pushing the zoom to gps position button (see figure 3.7. The map viewport

14

2. BeeGIS

Figure 2.14: The gps view containing the realtime information of the gps.

will immediately be centered on the gps position.

It is important to be aware of the fact that once gps logging is enabled every gps position is

stored in the internal database in background. These data can be useful for example to import a

set of pictures taken during the survey, that then can be aligned through their timestamp to the

positions of the gps point with the most similar timestamp.

2.2.4 Gps points acquisition tools

There are several way to capture data from the gps and put them on a feature layer. The first

thing to keep in mind is that every action is executed on the selected layer and is meant for the

type of geometry found on that layer.

Manual and automatic data acquisition

Two tools are dedicated to the addition of features on map layers, the manual and automatic

point addition.

The manual point addition tool, activated by the icon of figure 3.4, behaves differently depending

of the type geometry found on the selected layer at the time of pushing the button:

15

2. BeeGIS

• point layer - a point is added to the layer

• lines layer - a point is added to the last line feature found in the layer

• polygon layer - a point is added to the lat polygon feature found in the layer

The same applies in the case of the automatic point addition tool, activated through icon of

figure 3.6. The only thing that changes, is that in this case a process is started that places in

automatic mode points on the layer, following the time interval set in the gps settings panel.

The automatic and manual mode can be used concurrently, both on the same or on different

layers. The automatic acquisition mode is started on the selected layer and keeps in memory the

layer until the tool is not stopped. Hence, once started, it is possible to select other layers and use

the manual acquisition mode on them, without conflicting.

There are situations in which it is important be able to use both modes. One simple example

is the survey of a basketball field. The user would create a polygon layer, step to the border of

the field and start automatic data acquisition. Then he would put the tablet pc in a bag and walk

the edges of the field. Depending on the time and distance threshold chosen by the user in the gps

settings, he will observe different error amplitude. In fact, most probably the corners of the field

will have been smoothed, if the user didn’t stop on a corner in the exact moment. The solution is

to walk an edge to the corner, pick out the tablet pc and manually add the corner point.

Geonotes placed on gps positions

There is a third way to add gps data on the map, through the geonote addition tool, which is

triggered by the icon of figure 3.5.

In this case once pushed the button, a new geonote opens up, placed at the current gps position,

ready to be filled with informations. Geonotes will be discussed in the next section (3.3).

2.2.5 Layer creation aid tools

Since BeeGIS’s strategy is leans towards keeping things fast and simple for users out in the filed

survey, a quick way to create empty layers on which to work on is provided. Through one single

click with the pen on one of the three last icons on the gps toolbar (figures 3.8, 3.9 and 3.10) a new

layer of the selected geometry type is created with two default attribute fields. The only thing the

user has to define, is where to save the shapefile that will be generated.

16

2. BeeGIS

2.3 The Geonotes

Probably the most intuitive tool of BeeGIS is represented by the Geonotes. The PostitTMnotes

have been in every office for years now and everyone used them at least once to stick quickly drawn

notes on the pc monitor, in a book, everywhere necessary to be able to quickly recover from the

note the thought of the moment in which the note was written.

Geonotes work exactly the same way and they can be sticked on to maps and filled with any kind

of information, from fast drawing through the tablet’s pen, to text inserted through a keyboard,

to any kind of media and document.

2.3.1 Anatomy of a geonote

When a new geonote is created, its properties are set to the default values and it looks like figure

3.16.

Figure 2.16: A new created geonote.

The title bar, in the top part of the note, contains, several informations:

• the icon of the note, which also describes the origin of the note, which can be a normal note

created with the geonotes tool, or a note taken by gps or a note imported from a pictures

folder or from Geopaparazzi (see 4.4).

17

2. BeeGIS

Figure 2.17: The default geonotes icon.

Figure 2.18: The icon of geonotes placed through the gps tool.

Figure 2.19: The icon of geonotes created through pictures imports.

• the title of the geonote, which is by default generate by concatenation of the currently opened

map’s name, the current project name and the database id of the geonote.

Some of the properties of the geonote can be changed by double-clicking on the title bar, which

makes appear the properties panel of figure 3.20. The properties panel has a top information area

reporting timestamp and position coordinates. The middle are is dedicated to the editing of the

title and the geonote’s color. In the bottom part of the properties panel four buttons are provided

to accept the changes, cancel the changes, dump a geonote to disk and delete the current geonote.

The last two options will be better explained in the next sections.

Figure 2.20: The properties panel of a geonote.

18

2. BeeGIS

At the top left side of the geonote there are two icons that have both the result of closing the

geonote. The first icon from left saves the changes to the geonote to to disk, while the second

simply closes the geonote without saving anything.

The bottom part of the geonote is occupied by the colors bar. This is a small bar of buttons for

quick changing of the color of the geonote. While from the properties panel it is possible to change

to any color, here a set of predefined colors is provided. These colors are also used in the search

engine of other parts of the application.

The center part, or body, of the geonote is is made up by a set of tabs, representing the different

ways to insert informations into a geonote.

2.3.2 The Paint box

The first panel of the geonotes body, as shown in figure 3.21, gives the user the possibility to

draw his thoughts and interpretations of the environment freehand with the pen as he was drawing

on paper. The paint box toolbar provides the possibility to chose between a set of predefined

stroke thicknesses, a set of stroke transparencies and the color of the stroke can also be chosen.

Furthermore the drawn strokes can be removed through the last two icons on the toolbar. Since

geonotes are resizable, larger space for the sketches can easily be created, if necessary. The content

of the geonote can be exported as image by right-clicking in the middle of the paint area. A dialog

is proposed for saving the content of the box as an image to file.

Figure 2.21: The paint box of a geonote.

19

2. BeeGIS

2.3.3 The Text box

The second tab represents the text box, which is basically a text area into which the user can

write text through a keyboard, whether physical or on-screen (see figure 3.22).

Figure 2.22: The text box of a geonote.

2.3.4 The Media box

The third tab is the media box.This area can be seen as a storage area for any type of data.

Figure 3.23 shows an example of media area populated with very different types of files.

In fact the user is allowed to drag and drop any type of file from the filesystem (for example the

filemanager of windows) onto the media box, which will add the file to the list of contained data,

supply it with an icon if the mimetype is recognized, and store the file into the internal database.

The media box provides a content menu accessible through right-click on an item of the list or

in the middle of the box area. The menu proposes to delete a selected item or the complete clearing

of the area, in which case it is important to be aware of the fact that the media to which the items

are connected to, will also be removed from the database and can’t be recovered.

One more available option is the open with system viewer action, which opens the currently

selected item with the system editor associated to the mimetype. This means that for example an

item with extension .doc is opened hopefully with Openoffice1 or in the worst case with Microsoft

1http://www.openoffice.org

20

2. BeeGIS

Figure 2.23: The media box of a geonote.

WordTM, a .pdf file is opened with Acrobat Reader and an image of type png or jpg is opened with

the system image preview application.

The same happens in the case of a double-clicking on any item in the media box. all files apart

of the image files are opened with the default system editor. Images instead are opened with the

internal BeeGIS image editor, which will be described in the next section.

The image editor

The internal image editor is a very basic viewer, that provides certain editing and drawing

capabilities. An editor, opened on an image from the mediabox, can bee seen in figure 3.24.

The main toolbar provides tools for drawing with strokes of different width, transparency and

color on the image, as well as tools for removing all or single strokes from the image. Basic zooming

capabilities are also provided, like the zoom to fit the image to the editor, zoom in and zoom out.

By right-clicking on the image area, it is possible to export the image to disk together with the

drawn annotations.

It is important to know that once the image is saved through the save button at the bottom

of the editor, the image is saved into the database separately from the drawn annotations. This

means that the image is always kept in its original state and the drawing on the image can be

removed at any time. In fact, when opening the image with the system editor, the annotations

drawn by the user do not appear.

21

2. BeeGIS

Figure 2.24: The internal image editor of BeeGIS.

2.3.5 Geonotes tools

The geonotes toolbar provides two tools, the geonote tool and the selection tool. The first

is the most used tool, since through it new geonotes are created by simply clicking on the map

viewport, and old geonotes are opened, when found inside the selection box that the tool creates

when dragged over the map. If more than one geonote is inside the selection box, all of them are

opened.

Figure 2.25: The geonotes toolbar.

The selection tool is strictly related to the fieldbook, and will therefore be handled in the next

section (3.3.6).

22

2. BeeGIS

2.3.6 The fieldbook view

The possibility to stick a large amount of geonotes on maps raises also the need to properly

organize the geonotes and be able to search them properly, as well as browse them as if they were

pages of a paper fieldbook (BRINER ET AL., 1999).

BeeGIS’s fieldbook exactly serves to this purpose. It can be opened by clicking on the icon of

figure 3.26 in the main toolbar.

Figure 2.26: The fieldbook icon on the main toolbar.

The fieldbook is a view, that was separated into two parts, were the upper part lists all the

available geonotes and the lower part visualizes the geonote that was selected in the upper list.

Figure 3.27 shows the opened fieldbook inside BeeGIS, without any selection active.

Figure 2.27: The opened fieldbook in BeeGIS.

Search features

The fieldbook gives several ways to search for geonotes inside the book itself but also on the

map.

The most immediate way to search for geonotes on the map, is to select them in the fieldbook

23

2. BeeGIS

right-click over them and select the zoom to geonotes action. The map will immediately zoom to a

large enough maparea able to contain all the selected geonotes, highlighting the pins of the selected

geonotes (see figure 3.28).

Figure 2.28: Geonotes selected in the fieldbook are highlighted on the map.

Vice versa it is possible to select geonotes by dragging a bounding box on the map through the

geonotes selection tool of figure 3.25. In that case only the selected geonotes will be visible on the

fieldbook geonotes list. Figure 3.29 shows the selection of three notes.

Figure 2.29: Geonotes selected on the map through the selection tool, are filtered in the fieldbook panel.

The top of the fieldbook provides four other ways to filter the list of geonotes, that can be chosen

by selecting the mode in the available combobox.

By default the text search is active. In that mode, below the combobox there is a textarea. If

24

2. BeeGIS

the user inserts text in that area, that text is searched in the title and textbox of every geonote

and only geonotes matching that text are left in the list. In image 3.30 this is easy to understand,

since the searched text is in the name of the filtered geonotes.

Figure 2.30: The fieldbook in text search mode.

Another way to search for geonotes is the color search mode. In that case below the combobox

the colors of the geonote’s color toolbar are proposed as buttons. The selection of one of the color

filters out only the geonotes that have that particular color assigned. Figure 3.31 shows that the

user selected the blue button and that there are three geonotes with that color assigned in the

database.

The third way to search for geonotes, is by the selection of a timeframe inside which the geonote

is supposed to be created. In that case below the combobox three buttons appear, the first to define

the start date, the second one to define the end date of the timeframe, and the third button, that,

once the start and end date have been defined, filters out only those geonotes that were created

inside of the given timeframe (see figure 3.32).

The last way provided to search for geonotes, is the possibility to filter them by type. In that

case it is possible to select only those geonotes created through gps, or created by import of pictures

or only those geonotes created through the pen by the user (see figure 3.33).

25

2. BeeGIS

Figure 2.31: The fieldbook in color search mode.

Figure 2.32: The fieldbook in date search mode.

26

2. BeeGIS

Figure 2.33: The fieldbook in type search mode.

The context menu

It is possible to execute several actions from the fieldbook. Those are available through the

popup menu, that appear when right-clicking on the geonotes list panel. The full menu can be seen

in figure 3.34.

A description for every action is hereby supplied:

• Zoom to geonotes - zooms the map viewport to all the currently selected geonotes in the list.

• Remove geonotes - deletes all the currently selected geonotes from the map view and the

database.

• Export to shapefile - exports the geonotes informations contained in the title and textarea

to a point shapefile, with three attributes fields: title, text and timestamp of creation of the

note. The result of an export of this type is shown in figure 3.35.

• Dump geonotes - dumps the currently selected geonotes to disk in the most possible human

readable way. This means that a folder for every geonote and based on its title is created.

Inside that a folder containing all the items of the mediabox is generated, as well as a picture

27

2. BeeGIS

Figure 2.34: The context menu of the fieldbook and its features.

28

2. BeeGIS

containing the sketch of the paintbox and a file containing the text present in the textarea.

Also a file containing informations about the geonote, like title, description, creation date,

position and coordinate reference system definition is created.

• Dump geonotes binary - dumps the geonotes in a binary format, useful for import and export

of the geonotes. This dump generates a compressed archive containing the selected geonotes,

that can be given to any other BeeGIS user. That archive can then simply be dragged and

dropped into the geonotes list panel of the fieldbook of the other user and that way it will be

imported automatically into the fieldbook and database.

• Import geonotes archive - imports a geonotes archive created with the binary geonotes dump.

• Send geonotes - exports the currently selected geonotes and sends them via mail to a prede-

fined email. To make this working properly, it is necessary to configure the outgoing email

informations properly in the main preferences panel under the item geonotes (see figure 3.36).

Once triggered, BeeGIS generates and email, attaches the exported geonotes archive and sends

it to the predefined email address (see 3.37). The attached archive can then be imported or

dragged into the fieldbook.

Figure 2.35: Geonotes exported to a point shapefile that contains informations about title, text and creation times-tamp.

29

2. BeeGIS

Figure 2.36: The preference panel of BeeGIS, in which the configuration for the outgoing email server is set, in orderto be able to send geonotes via email.

Figure 2.37: The email generated by BeeGIS to send geonotes. The attachment of the email is the geonotes archivethat can be imported into the fieldbook.

30

2. BeeGIS

2.4 The Annotation Layer

Where the traditional way of field mapping uses colored pencils on a paper map, BeeGIS offers

the annotation tool. It can be activated through the icon of figure 3.38.

Figure 2.38: The annotation tool icon in the main toolbar.

The annotation tool allows to draw lines and fill areas on the map (see figure 3.39).

Figure 2.39: Annotations drawn over the map.

Once activated, the annotation tool itself as well as the annotation view is opened (see figure

3.40), which contains the drop down menus to select stroke thickness, color and transparency of the

used pencil. There are also two other buttons that can be used to clear all the annotations from

the map and to delete only the last inserted annotation from the map, without having to change

tool.

in the annotations toolbar, shown in figure 3.41 a second tool is provided that can be used to

remove in a more easy way the annotations.

Because we are supposed to be on the field with a tablet where is not so easy to pick exactly

31

2. BeeGIS

Figure 2.40: The annotation view, with the dropdown menus for defining the pencil’s properties.

Figure 2.41: The annotations toolbar.

32

2. BeeGIS

the stroke we want to remove, the remove annotations tool is not a pick tool, but instead a line

tool. When dragging the tool over the map a red straight line is drawn and all the intersecting

annotations are removed as soon as the tool is released. Figure 3.42 shows the red line drawn by

the remove annotation tool an instant before the pen is released from the screen. Once released,

all the blue annotations are removed.

Figure 2.42: The annotations remove tool in action. The red line crossing the blue ones is the stroke of the removetool. Once the tool is released, all the intersected annotations are removed from the map.

2.5 The embedded database

BeeGIS adds to the standard data used in GIS several informations. For most of these infor-

mations it was not possible to embed them into one of those GIS formats. Being one of the first

aims of the author to follow the standards, many efforts have been put into import and export

tools from and towards the standard formats. This happens for example in the case of exporting

geonotes layers to shapefile (see 3.3.6) or Geopaparazzi data to KML (see 4.3.1).

To store all data that do not fit in any of the standard GIS data formats, the choice was made

to use an embedded database for BeeGIS and persist any data from BeeGIS. Therefore for example

Geonotes, as well as media saved in the Geonotes, data from the annotation layer are stored in the

33

2. BeeGIS

internal database.

The choice of the embedded database fell on the H22 database for several reasons:

• language - H2 is written in Java and therefore is particularly suitable for working together

with BeeGIS.

• portability - being H2 written in Java, it works properly on all the operating systems on which

BeeGIS can be used.

• local webserver - H2 comes with a webserver and a graphical user interface which is accessible

via browser and can therefore be used inside BeeGIS to query data in an advanced and manual

way.

• spatial extensions - H2 has some features for storage of spatial features.

2.5.1 H2 database

The H2 embedded database, once configured and started, creates a couple of files inside the

folder that was supplied as the root folder for the database. If the database already exists, the

existing one is opened, if it doesn’t exist, a new empty one is created.

The configuration of the database can be done in BeeGIS’s main preferences, under the Embedded

Database Server item. It requires to choose:

• the database type, which currently can be only H2 database.

• the tcp port to use for the connection to the database.

• the database user, which by default can be sa.

• the database password, which can be left empty, in the case of the default sa user.

• the database folder, which is the folder into which the database sill be created, if not existing.

By default this folder is set to an internal folder in the BeeGIS installation.

The main preferences panel is shown in figure 3.43.

Since the database is completely contained in one folder, it is quite easy to exchange BeeGIS

databases between users. It is as simple as copying the database folder onto the other user’s

computer and redirect the database folder in the preferences panel (figure 3.43) to that folder.

Once BeeGIS has been restarted it loads automatically all the geonotes and annotations that were

contained in the database.

The content of an existing database folder of BeeGIS is shown in figure 3.44.

2http://www.h2database.com

34

2. BeeGIS

Figure 2.43: The database settings panel in the main preferences.

Figure 2.44: The content of an H2 database folder. in this case moving the database of BeeGIS is as easy as copyingover to another computer the databasebeegis folder.

35

2. BeeGIS

2.5.2 The embedded database view

How geonotes and annotations and any other information of BeeGIS is stored to the database

is kept hidden to the average user on purpose. It is possible anyway that some advanced user may

want to access the database and query it the manual way. BeeGIS, through the tools of H2, gives

the possibility to access its internal database via a browser user interface, which is embedded into

a view. To access the database view, it is necessary to open the window menu (figure 3.45), from

there click on the other entry (figure 3.46) where the Embedded Database View can be selected.

Figure 2.45: The window menu, the starting point to get to the database view.

Figure 2.46: The views menu, from where the embedded database view can be opened.

36

2. BeeGIS

The view that is opened in the lower part of BeeGIS is shown in figure 3.47 and is very intuitive

user interface, which even provides some handy tools like sql autocompletion.

Figure 2.47: The database view, which opens directly on the active embedded database connection.

The Embedded Database View really serves as a database client, inside which any sql query can

be executed. Also data can be edited, deleted or inserted.

37

2. BeeGIS

2.5.3 The database structure

The database structure of BeeGIS is shown in figure 3.48. The tables have been split to be able

to optimize the extraction of basic data for geonotes, while the internal data as for example media

files are extracted once they are needed.

Figure 2.48: The graph of the database tables used in BeeGIS.

To access the database, the Hibernate3 library has been used. Hibernate gives the possi-

bility to handle the database at a more clean level, dealing directly with the database tables

through the classes in the code that were mapped to database tables (see [Bauer et al., 2007] and

[McKenzie, 2007]). That way the code gets much cleaner and better readable than in the case of

directly accessing the database through the libraries provided by the java language (JDBC4).

3http://www.hibernate.org4http://java.sun.com/javase/technologies/database/

38

2. BeeGIS

2.6 Other tools

2.6.1 Gps data logs export

BeeGIS provides the possibility to export the internal gps log, i.e. the log that is created every

time the gps logging is enabled (see section 3.2.3), to shapefile.

The export tool is accessible from the general export action in the file menu (see figure 3.49).

Figure 2.49: The export wizard with the gps log export entry.

Once the Gps export tool is selected in the export wizard, the panel of figure 3.50 is displayed.

In this panel it is possible to supply a start and end date inside of which one wants to extract the

gps points. If the dates textboxes are left empty, the whole log is extracted.

It is also possible to select whether to export the log as layer of lines or points. Once the output

path for the new created shapefile is supplied, it is possible to push finish. Figure 3.51 shows an

example layer created from a complete gps log export.

Figures 3.52 and 3.53 show the case in which the log is exported to a lines layer, but limited to

the points created between the timestamps 2009-07-03 09:57 and 2009-07-03 09:58.

39

2. BeeGIS

Figure 2.50: The gps log export wizard, filled to export the whole gps log to a point shapefile.

Figure 2.51: A gps log exported to a point shapefile.

40

2. BeeGIS

Figure 2.52: The gps log export wizard, filled to export the the gps points created between two dates to a lineshapefile.

Figure 2.53: A gps log exported to a line shapefile.

41

2. BeeGIS

2.6.2 Photo import

On a survey it is important to be able to connect the measurements part with pictures taken

during the survey, particularly when one is doing the postprocessing of the collected data. As

already stated before, BeeGIS’s aim is to keep the postprocessing as low as possible, in order to

limit the possibility of errors in transcription, re-digitalization and interpretation.

BeeGIS provides a tool dedicated to the alignment of pictures taken with a digital camera with

the internal gps log. The tool can be best explained through an example. The following example

has been done on a linux computer, which explains the reason for the different window appearance

in the following figures.

The following example will assume that a survey has been done and that the user took several

gps drawn lines and geonotes, as well as several pictures with a digital camera.

Once the user starts his postprocessing, he will be in the situation shown in figure 3.54, i.e.

visualizing the taken geonotes, gps paths drawn and annotations.

Figure 2.54: The start situation for the import of pictures, after a survey.

What is not visible in figure 3.54 is the background gps log that was taken during the survey.

It is possible to visualize the data in the gps log by using the embedded database view (see section

3.5). For example to have an idea of how many gps points were taken in the log, it is possible to

execute the query SELECT count(*) FROM GPSLOG inside the database view as is shown

42

2. BeeGIS

in figure 3.55.

Figure 2.55: The database view and a query that counts the available gps log points.

The user will have to connect the digital camera he used for the surveying to the computer and

will then be able to visualize the folder with the taken pictures, in the path where the operating sys-

tem mounted the camera’s storage media. In the case of figure 3.56 the camera was mounted on the

folder /media/disk and the pictures taken are inside the subfolder /media/disk/DCIM/100CANON.

After these preparation parts, it is possible to start the import wizard from the main file menu

under import. The generic import wizard will show up as in figure 3.57 and propose an Import

Photos entry.

Once selected the Import Photo, the panel to import the pictures is visualized. The panel need

just two informations, that are:

1. the folder that contains the pictures to be imported. It is very important to note that

BeeGIS is not able to read the creation time of a picture, but only the last modification time.

Therefore it is necessary to execute the import directly on the camera mounted on a local

folder. If the imaged are instead copied to disk, this will change the the last modification

time and therefore make the import tool useless.

2. the time shift between the gps time (UTC) and the camera time at the time of shooting the

pictures. This value defines the accuracy of the import and obviously needs to be checked

when starting to take pictures during the survey. At that point it is simple to compare the

43

2. BeeGIS

Figure 2.56: The list of pictures in the camera.

Figure 2.57: The import wizard of BeeGIS, where the photo import entry is available.

44

2. BeeGIS

gps time visible in the gps view of BeeGIS and the camera time.

Figure 3.58 shows the import panel filled with example data.

Figure 2.58: The photo import wizard.

Once finished is pushed, BeeGIS starts the importing process. During the process the pictures

timestamp are compared with the internal gps log points and in the position of the nearest gps

point by timestamp a new geonote is created, containing the picture in the geonote’s mediabox.

Pictures that have no gps point taken at the time of shooting, are ignored and at the end of the

import a list of not imported pictures is proposed to the user to be able to understand if the import

has been successful.

Figure 3.59 shows the situation after a successful import of pictures. BeeGIS in that case shows

a geonote, named after the imported picture, for every imported picture. in the figure it can be

noted that in the mediabox the picture is available for further processing.

2.6.3 Geopaparazzi import

BeeGIS provides tools to import data from the lightweight surveying solution Geopaparazzi into

the BeeGIS workspace, creating geonotes and shapefiles from the imported data. The import tool

will be discussed after the description of the Geopaparazzi application, in section 4.4.

45

2. BeeGIS

Figure 2.59: The imported pictures wrapped into geonotes. The icon shows that the geonote contains an importedimage.

46

3. A lightweight solution for fast surveys

3 A lightweight solution for fast surveys

Depending on the situation, it is possible that a lightweight solution for surveying might be more

appropriate. A lightweight solution might be based on a smartphone for example, which makes

such a tool always present in anyone’s travel bag.

Since the advent of the IPhoneTMby AppleTMmany applications that were meant to be used on

pcs before, found their way into the hands of anyone.

The response of GoogleTMto the IPhone didn’t take to long and the AndroidTMplatform1 was

born. The main difference between the two is probably the fact that in the case of the IPhone,

following Apple’s strategy since ever, both the hardware and the software are developed by the same

party. So IPhone really stands for a complete hardware and software system, whereas Android,

being an operating system, represents only the software part.

The choice for the development of the lightweight survey solution fell almost immediately on

the Android platform for the following main reasons:

• the Android platform was released by Google under open source license. The source code

of the platform itself is available and can be used. This assures complete freedom over the

developed application.

• the Android platform is developed in JavaTMand as such no big gap was formed between the

BeeGIS and the Geopaparazzi development, leaving the possibility to code exchange between

the two projects open.

3.1 The Android platform

Even if the Android platform was started by Google, it was developed by several of the most

important technology companies, that form the Open Handset Alliance2 (OHA). The OHA is

currently a group of 47 technology and mobile companies3 who, as their mission states, have come

together to accelerate innovation in mobile and offer consumers a richer, less expensive, and better

mobile experience. Together they have developed Android, the first complete, open, and free mobile

platform. The OHA is committed to commercially deploy handsets and services using the Android

Platform[Ableson et al., 2007].

1http://www.android.com2http://www.openhandsetalliance.com3Examples are: Sony Ericsson, Samsung, HTC, Telecom Italia and Toshiba.

47

3. A lightweight solution for fast surveys

In fact the Android platform was developed in order to make the applications development

process easier and therefore faster.

Apart of its clear and accurate documentation it ships as a software development kit (SDK) and

several tools that integrate with the most important Integrated Development Environments (IDE)

as can be seen in figure 4.1.

Figure 3.1: Android application development with the Eclipse IDE. The plugins ship with a complete emulatorenvironment. Once the phone is connected, the IDE prompts the user on which device to run the application.

The IDE also provides a deployment mechanism, in order to be able to package the application

to be installed easily on other devices.

3.1.1 Hardware features

The first android enabled phone has been the G1 DreamTMby HTC. Since then several other

companies started shipping their android enabled phones. Being the oldest Android enabled phone

it will be taken to describe the minimum hardware features available:

• 3.17” touchscreen display (480 x 320)

• 3.0 megapixel camera

• Bluetooth 2.0 + Enhanced Data Rate

• GPS

• compass

48

3. A lightweight solution for fast surveys

• accelerometer

For the Geopaparazzi application, the minimum requirement was the presence of a GPS and the

compass, which both were fulfilled by the G1 Dream.

3.2 Geopaparazzi

The main aim of Geopaparazzi was to create a tool that:

• would fix in any pocket and be always at hand, when needed.

• would give the possibility to take georeferenced and possibly orientated pictures during the

survey, with further possibility to import them into the main GIS application BeeGIS.

• would be able to exploit easily internet connection, if available.

• would be extremely easy to use and intuitive, providing just few important functionalities.

The main features available in Geopaparazzi are:

• georeferenced notes (section 4.2.1)

• georeferenced pictures (section 4.2.2)

• gps logging (section 4.2.3)

• easy export of collected data (section 4.3)

• a map view for the navigation of the environment (section 4.2.4)

All the features that need to be quickly accessed, as toggling gps on and off, create a note or

take a picture, as well as visualizing the current position on a map, are accessible from the main

view of the application, as can be noted in figure 4.2.

The main view presents a compass and main gps information, as well as four large and easy

accessible buttons.

It is important to note that most of the smartphones have the possibility to work in portrait

and landscape mode. This is most useful in the case of taking a picture, but is often necessary.

One example is the case of the G1 Dream, that has a real keyboard which, whenever used, forces

the phone to work in landscape mode, as figure 4.3 shows.

The possibility to have both landscape and portrait mode on the phone, made it necessary to

develop several checks in order to have a proper button displacement for enhanced readability,

49

3. A lightweight solution for fast surveys

Figure 3.2: The geopaparazzi application as it shows up when it is started.

Figure 3.3: The HTC G1 Dream in keyboard usage mode.

50

3. A lightweight solution for fast surveys

but also, which is even more important, to have the compass properly working, since the internal

coordinate reference system switches axes system for the different orientation modes. An example

of the Geopaparazzi’s main view in landscape mode can be seen in figure 4.4.

Figure 3.4: The geopaparazzi view in the case of horizontal disposal of the phone.

If the first access level for the most important features is provided by the easy accessible buttons

on the main view, the second access level on every view is represented by the view’s menu, which

can be accessed through the menu button every phone provides. The view’s menu has the powerful

possibility to give access to several functionalities without leaving the current view.

Geopaparazzi’s main view’s menu, as shown in figure 4.5, provides several functionalities as for

example the export to KML function (section 4.3.1), which gives the possibility to visualize all the

collected data during the survey in the Google EarthTM3D visualization environment4.

One of the most similar feature with BeeGIS is for sure the possibility to take a georeferenced

note. While in BeeGIS this gives also the possibility to draw and add media, in the case of

Geopaparazzi this is not possible, or better, it has been split into two:

• the textual part: Georeferenced notes (section 4.2.1)

• the images part: Georeferenced pictures (section 4.2.2)

In Geopaparazzi it is not possible to draw on notes.

3.2.1 Georeferenced notes

On Geopaparazzi the note is really a textual note and every note will be saved as a separate

textfile to the phone’s storage media, together with the information of the position in lat/lon

4http://earth.google.com

51

3. A lightweight solution for fast surveys

Figure 3.5: The main view’s menu.

coordinates and the timestamp in utc format. The note is saved in a human readable format and

the content of note on disk can be visualized with the phone’s own textfile viewer, as figure 4.19

shows.

Once the button take a gps note is pushed, the note view is presented to the user, which prompts

the user to insert any text in a textarea (see figure 4.6). Once the ok button is pushed, the content

of the textarea is made persistent on the phone’s storage media.

The notes are saved with the name containing the timestamp following the pattern:

note YYYYMMDD HHMMSS.txt. More on this will be covered in the data structure section (4.3.2),

the content of a note looks like the following:

utctime=2009-06-15 08:52

lat=46.681283712387085

lon=11.134455800056458

text=home sweet home

52

3. A lightweight solution for fast surveys

Figure 3.6: The view that opens for the insertion of a new note.

3.2.2 Georeferenced pictures

It is clear that a smartphone camera can’t compete by any means with professional cameras

and that is also not the purpose in Geopaparazzi. But in the case of a lightweight survey every

quick information can be of value and nowadays every smartphone provides a camera, so why not

exploiting it to take fast shots of the environment?

The take picture button on the main view takes the user to the well known camera preview view,

which is the same view that the system accesses when a user takes a picture with the phone the

usual way. The only difference is that in the Geopaparazzi camera preview, when the user takes

the picture, two files are saved to disk:

1. the image taken by the camera, with the name containing the date following the format:

IMG YYYMMDD HHMMSS.jpg

2. an information file for the image, saved with the same name but different extension: *.prop-

erties. The file contains informations that will be used by BeeGIS when importing them:

• longitude

• latitude

53

3. A lightweight solution for fast surveys

• azimuth

• altimetry

• timestamp in utc

An example information file looks like the following:

latitude=46.68165385723114

longitude=11.134788393974304

azimuth=1

altim=396.0

utctimestamp=20090825_105427

Before deciding to create a companion file for every taken picture, the possibility was investigated

to use Exchangeable image file format (Exif5) specification used by digital cameras to store the

needed informations through tags inside of the image itself. This was achieved up to a certain point

but subsequently was discarded because too resources absorbing. In fact the available libraries in

java to manage Exif tags would have required to be adapted to work on Android.

The compass

One important feature of Geopaparazzi is the fact that it stores in the picture information file

also the orientation in degrees, with which the picture has been taken.

To correctly obtain this value from the phone’s compass, the internal coordinate system had to

be remapped. This requires further explanation.

As figure 4.7 shows, the original orientation of the phone’s internal compass is mapped with the

Z axis pointing in direction outside of the phone’s screen. The azimuth angle that the compass

supplies in this case can be used when the phone is hold in horizontal mode as for example with

the back laid on a table.

In Geopaparazzi the aim is to exploit the compass to gather the planar angle whenever taking a

picture, in order to exploit properly BeeGIS’s capability to orientate geonotes on a map (see 4.4).

Basically the compass should properly measure azimuth angles when the user keeps the phone

in landscape mode with the screen in the direction of the user’s face (as visualized in images 4.7

and 4.8).

To achieve the above requirement, the axes of the compass were remapped in order to swap the

X axis with the Z axis. A visual representation of the change can be seen in figure 4.8).

5http://en.wikipedia.org/wiki/Exif

54

3. A lightweight solution for fast surveys

Figure 3.7: The axes orientation of the android phone. While taking a picture the angle around the X axis isinterpreted as the azimuth, which is wrong.

Figure 3.8: The remapped axes orientation of the android phone. The Z axis, which supplies azimuth values, nowpoints in the right direction.

55

3. A lightweight solution for fast surveys

Given the above changes, the compass view will properly interpret the azimuth angle when the

user keeps the phone in front of him simulating to take a picture. Figure 4.9 shows two examples of

orientation of the phone (which looks in the same direction of the user) and the triggered azimuth

angle in the compass view.

Figure 3.9: The orientation of the compass needle in the case in which the user looks more or less in the direction ofthe north and in the case in which the user looks slightly to his right.

3.2.3 Gps logging

The start gps logging button is the only toggle button, i.e. a button that gives the possibility to

enable and disable an activity. This is the needed behavior since gps logging is done in background

without any interaction with and without blocking the application.

When the logging button is pushed, the button changes color and the text below the button

states stop gps logging to advise the user that gps logging is enabled (see figure 4.10).

56

3. A lightweight solution for fast surveys

Figure 3.10: The gps logging toggle button changes color when it is recording gps positions in background.

Once gps is enabled, a new gps log file is saved with the name containing the timestamp following

the format:

GPSLOG YYYMMDD HHMMSS.txt

The file contains one gps acquisition per line, in comma separated format: longitude, latitude,

altimetry, timestamp.

If the gps logging is stopped, the file is closed and all the data contained in that gps log file are

assumed to be part of one line entity. That means that if the gps logging is started again, a new

gps log file is created and all acquired data until next logging interruption are appended to that

file.

A sample gps logging file looks like the following:

10.940912961959839,45.735434889793396,183.0,2009-06-16 08:07:45

10.940923690795898,45.735429525375366,184.0,2009-06-16 08:08:00

10.940934419631958,45.73541879653931,183.0,2009-06-16 08:08:15

10.940934419631958,45.73541879653931,183.0,2009-06-16 08:08:30

10.940945148468018,45.73541343212128,185.0,2009-06-16 08:08:45

10.940945148468018,45.73541343212128,186.0,2009-06-16 08:09:00

57

3. A lightweight solution for fast surveys

3.2.4 The OpenStreetMap view

The show position on map button takes the user from the main application view to the Open-

StreetMap6 view, which will be referred to simply as the map view (as opposed to the google maps

view).

OpenStreetMap (OSM) is the project of a free editable map of the whole world. It is based on

the collaborative insertion of geographic information all over the world by any person interested.

As their mission states, ”the project was started because most maps you think of as free actually

have legal or technical restrictions on their use, holding back people from using them in creative,

productive, or unexpected ways”.

Even if OSM is an incredibly important projects, describing it is beyond the intent of this thesis.

Interested readers might want to browse the available informations on the main wiki site7.

For the purposes of Geopaparazzi it is enough to say that OSM gives the possibility to download

maps for visualization of the current gps position. Differently from its proprietary counterpart

Google Maps, with OSM it is also possible to retrieve maps when internet is available, cache them

on disk and use them from that moment on even when in offline mode.

The map view, when first opened places itself on centered on the current gps position if available.

Fetching tiles from OpenStreetMap in online and offline mode

The map tiles fetching mechanism works at three different levels in Geopaparazzi. Once the

position to show is defined, the application will try to fetch the map tiles to be shown as follows:

1. retrieve them from memory. The application provides an in-memory tile cache in which a

certain amount of tiles are kept in memory for extreme quick access. When tiles from memory

are used, the panning and zooming over the map will occur completely smooth.

2. if the requested tile isn’t available in the tile cache, the application checks if the tile is available

inside the on-disk tile cache, which is created every time a tile has to be fetched online. If the

tile is available on disk, it is read and rendered and also pushed into the in-memory. If the

in-memory tile cache reaches it’s limit, for every new tile inserted, the oldest available tile is

removed from the cache and its memory is released.

3. if the requested tile isn’t available also in the on-disk cache, the application tries to connect

6http://www.openstreetmap.org7http://wiki.openstreetmap.org/wiki/Main Page

58

3. A lightweight solution for fast surveys

Figure 3.11: The Openstreetmap view, centered on the current gps position.

to the internet and to download the tiles directly from the OpenStreetMap tiles server. Once

fetched, the tile is saved in the on-disk cache and put in the in-memory cache. If no internet

connection is available, an empty crossed image is generated to show the user that the tile

could not be retrieved.

The tiling mechanism from the OpenStreetMap server needs some further explanation. In the

OSM logic the world is divided into tiles at different zoomlevels. The 0 zoomlevel covers the whole

world with one single tile, while the 1 zoomlevel splits the world into 2x2 tiles. Table 4.1 describes

the behavior for different zoomlevels.

0 1 tile covers whole world

1 2 x 2 tiles

2 4 x 4 tiles

...

n 2n x 2n tiles

Table 3.1: OSM zoom levels and the division in tiles.

59

3. A lightweight solution for fast surveys

The OpenStreetMap website8 supplies the necessary equations to be able to gather the needed

tiles given a position as longitude and latitude.

n = 2zoom (3.1)

xtile =lon deg + 180

360∗ n (3.2)

ytile =1 − log(tan(lat rad))+sec(lat rad)

π

2∗ n (3.3)

lon deg =xtile

n∗ 360.0 − 180.0 (3.4)

lat deg = arctan(sinh(π ∗ (1 − 2 ∗ ytilen

))) ∗ 180.0

π(3.5)

Given that every tile is downloaded as a 256x256 pixels image and that the smartphone screen

is 320x480 pixels, to make sure that the whole screen is covered by tiles, it is necessary to fetch 9

tiles, i.e. the tile containing the center coordinate of the screen and the 8 tiles adjacent to that.

The steps to perform to properly render the map are:

1. define the coordinate of the screen center. In our case this is the gps position.

2. through the center position it is possible to retrieve tile number (row and column) for the

actual zoom level that covers that position through the equations 4.2 and 4.3.

3. calculate the world boundaries of the interested tiles through the equations 4.4 and 4.5 and

the tile on screen boundaries given that the pixel size of a tile is known.

4. render the tiles at the proper positions on the map view’s canvas.

Every time the zoomlevel changes, the calculus has to be done again.

Navigate through coordinates and geocoding

Geopaparazzi provides several ways to navigate the map view. The most usual are probably the

panning and zooming.

Panning can be easily performed by dragging the map with the finger over the screen.

For zooming instead it is necessary to access the view’s menu through its dedicated button as

shown in figure 4.12.

There is also the possibility to navigate the map view by insertion either of coordinates of the

place of interest (POI) or by insertion of an address string.

8http://wiki.openstreetmap.org/index.php/Slippy map tilenames

60

3. A lightweight solution for fast surveys

Figure 3.12: The map view menu.

To do so, it is necessary to select the Goto button of the map menu, which will open a dedicated

view for the input of coordinates or addresses (see figure 4.13.

The navigation by coordinate is very simple to use. By inserting the coordinates of interest and

pushing the use coord button, the map view is centered on the position and visualized.

The navigation by address works in a different way and bases on the concept of geocoding.

Geocoding is the process of finding associated geographic coordinates (often expressed as latitude

and longitude) from other geographic data, such as street addresses, or postal codes9. In the case

of Android the supplied geocoder is the google maps geocoder and it needs an internet connection

to properly work.

This is the main difference between the two systems. The geocoding option is much easier to

use, but requires internet connection, whereas the coordinates option can be used always.

The result of both option is the same, i.e. a POI on which the map view is centered and

visualized (see figure 4.14). To be able to tell the difference between the gps position two different

icons where chosen, as table 4.2 shows.

9http://en.wikipedia.org/wiki/Geocoding

61

3. A lightweight solution for fast surveys

Figure 3.13: The dialog that permits to use both coordinates or a well defined address string to navigate the map.

Figure 3.14: The little green dot that represents the goto position, as opposed to the blue dot of the gps position.

62

3. A lightweight solution for fast surveys

the gps position icon the goto position icon

Table 3.2: The different icons to identify gps and goto position.

Google Maps

While the OSM map view is incredibly useful because of its possibility to work in offline mode,

the google map view is definitely more feature rich. It is important to keep in mind that it works

only when connected to internet, but it provides for example definition of routes, saving of places

and almost every feature available on Google maps10.

When selecting the googleview button from the main menu of the main view of Geopaparazzi,

the google maps view is opened on the current gps position, as shown in figure 4.15.

Figure 3.15: The google maps view, which also supports satellite data, but needs an internet connection to work.

10http://maps.google.com/

63

3. A lightweight solution for fast surveys

3.3 Export features

All data collected with the Geopaparazzi application are stored in a human readable format in

order to facilitate their import into different applications, were the data are meant to be processed

after the survey. In the case of BeeGIS a dedicated tool was created to best exploit all the informa-

tions gathered during the survey. While the import to BeeGIS is handled in section 4.4, the next

two sections explain the other possibilities to export the collected data from the phone.

3.3.1 Kml Export for Google Earth

The kml format was first known through the software that brought the 3D visualization of the

earth to almost any pc owner: Google Earth. The format also made it into the Open Geospatial

Consortium (OGC)11 getting a standard format.

It is possible to export the Geopaparazzi data to KML format pushing on the kml export button

of the main menu of the main view.

The export creates a KML file on the phone’s media storage that contains the informations of

the notes and gps logs:

• the notes are created as pins marks containing the information of the note text as title and

the timestamp as description.

• every log gps file file is taken as a line.

Figure 4.16 shows an example of Geopaparazzi content exported to KML format and opened in

Google Earth.

3.3.2 The disk data structure

Geopaparazzi uses for storage purposes the memory card that comes with every phone, in order

to make sure that the phone memory doesn’t get filled with data (for example OSM tiles), which

would probably block the phone and require expertise to be solved.

Once started the application creates a root folder named geopaparazzi and inside of that folder

the main data folder structure is created (see 4.17):

gpslogs the folder into which all the gps log files are stored.

kmlexport the folder into which exported KML files are stored.

11http://www.opengeospatial.org

64

3. A lightweight solution for fast surveys

Figure 3.16: An kml file exported from geopaparazzi and opened with google earth. The gps logs are visualized aswhite lines and the notes as red pins.

notes the folder into which the files containing notes are stored (see 4.18).

osmcache the folder used by the application to store downloaded OSM tiles (the on-disk tile

cache).

pictures the folder into which the pictures taken with the camera and their information files are

stored.

3.4 Upgrading to the desktop GIS, importing to BeeGIS

Geopaparazzi was designed to be lightweight and fast to use, but it is important to keep in mind

that it is a tool meant for surveying, which means that the collected data have a particular purpose

and need to be further processed. This can be done by importing the content of Geopaparazzi’s

data folder into BeeGIS.

BeeGIS has a import feature dedicated to Geopaparazzi data. Before starting any import

procedure it is necessary to connect the phone to the pc. By doing that the phone proposes to

mount the internal storage card as disk. It is mandatory to agree with that. Once mounted, the

phone’s storage can be browsed with the operating system file browser. At that point the device is

ready to be used from importing purposes.

Pushing the import action from the file menu (figure 4.20), several data types are proposed.

65

3. A lightweight solution for fast surveys

Figure 3.17: The root folder of the geopaparazzi application.

Figure 3.18: The folder content of the notes folder.

66

3. A lightweight solution for fast surveys

Figure 3.19: The content of a note.

Figure 3.20: The file menu and the import action.

67

3. A lightweight solution for fast surveys

By choosing the Import Geopaparazzi Data Folder type (figure 4.21) an import wizard appears,

which requires to insert the Geopaparazzi root folder, which was explained in section 4.3.2 and also

an output folder, into which generated shapefiles will be saved. An example configuration of root

folder and output folder are presented in figures 4.23 and 4.22.

Figure 3.21: The import wizard displaying the Geopaparazzi data type.

Figure 3.22: The wizard page properly compiled with input and output folders.

After pushing the finish button and once the import procedure has finished, some new items

will have been created and added to the map view of BeeGIS (see figure 4.24):

• notes - the shapefile layer created from Geopaparazzi’s notes folder.

• gpslines and gpspoints - the shapefile layer created from Geopaparazzi’s gps logs.

68

3. A lightweight solution for fast surveys

Figure 3.23: The selection of the Geopaparazzi root folder, once the phone has been mounted as disk H:.

Figure 3.24: A fresh import of Geopaparazzi’s data: gpslines and gpspoints, notes and geonotes.

69

3. A lightweight solution for fast surveys

• a geonote for every picture found in the pictures folder.

3.4.1 Notes import

Notes are imported a points layer with two attributes. The DESCRIPTION attribute into which

the note’s text is put and the TIMESTAMP attribute, into which the note’s timestamp is put (see

figure 4.25).

Figure 3.25: Notes are imported as a shapefile with a description and a timestamp attribute.

3.4.2 Gps logs import

Gps logs are imported as twice, once as line layer with two attributes, representing the timestamp

of the first point taken and the last point taken, and once as points layer, with just the timestamp

attribute (see figure 4.26).

3.4.3 Pictures import

Pictures are imported in a different way of notes and gps log. Since geonotes give the possibility

to load multimedia files, for every imported picture a new geonote is created and the picture is

stored inside of the geonote’s mediabox.

70

3. A lightweight solution for fast surveys

Figure 3.26: Notes are imported as a shapefile with a description and a timestamp attribute.

Since the pictures imported from Geopaparazzi hold also the information of the orientation, if an

azimuth angle is present, the geonote icon on the geonotes layer will show a little arrow describing

the orientation in which the picture has been taken.

In figure 4.27 three pictures were taken with Geopaparazzi standing more or less in the same

position. After every picture, the user rotated a bit on his left and then again took another picture.

The figure perfectly displays this behavior through the three little arrows bound to the geonote

icon. In the figure the three geonotes were opened and so the images in their mediaboxes. The

opened pictures were disposed in order to to easily recognize the motion of the user while taking

the pictures.

71

3. A lightweight solution for fast surveys

Figure 3.27: For every imported picture a geonote is created and the picture is stored inside the mediabox of thegeonote.

72

4. Use case: verification and update of datasets needed for the evaluation of the environmental impact of theproduction of hydro-electricity

4 Use case: verification and update of datasetsneeded for the evaluation of the environ-mental impact of the production of hydro-electricity

This chapter proposes a simplified description of a survey in which BeeGIS and Geopaparazzi

have been used to collect, correct and update data related to human works in the field of hydro-

electricity.

4.1 The objectives of the survey

The described survey has the objective to evaluate the environmental impact of the production

of hydro-electric power in the Trentino region in the northern part of Italy. The region, situated in

the heart of the Dolomites, has a dense distribution of human works for the hydraulics household

and hydraulic protection like dams, offtakes and artificial channels.

Many of the artifacts of interest were built before the sixties. For most of those the available

data have never been digitalized and are in fact available only in the form of the original designs

made by the building engineers. In those cases in which the paper data were digitalized, simple

shapefiles were drawn based on the technical maps and the project plans. This procedures have

introduced several incongruences between the digital data and the real world location of the works.

While in some cases the imprecision is a simple matter of wrong position of 10 to 50 meters, in other

cases the data are even missing or the difference to the real work is such to introduce incoherences.

In this last case it is necessary to do a field survey to verify all the available data. This means

that the digitalized data have to be compared with the original plans and also validated with a

survey.

The following survey has been done in the southern part of the Trentino region, around the

villages of Terme di Comano and Stenico. Figure 5.1 shows the area of interest.

4.2 Preparation of the cartography for the survey

One of the most important activities to gain a successful survey, is a proper preparation of the

cartographic base data needed during the survey.

73

4. Use case: verification and update of datasets needed for the evaluation of the environmental impact of theproduction of hydro-electricity

Figure 4.1: The region of interest, that was objective of the survey.

4.2.1 Raster cartography

The raster imagery data available for the survey were:

• a set of tiff data representing the technical maps of the region in WGS84/UTM ZONE32N

projection.

• a set of ecw data covering the same region as the technical maps and representing the or-

tophotos of the area of interest. The projection was the same as the one of the technical

maps.

Since the projection of the two datasets was the same, few had to be done to prepare them for

the survey. An operation of mosaic creation was performed, in order to create a single file that

loads all the visible tiles that compose the area of interest as if it was on single file. This operation

simply gains user comfort, since one has to deal with one file for the technical map and one for the

ortophoto, instead of having to load a whole pile of map tiles one by one.

4.2.2 Vector cartography

The vector cartography is more difficult to handle than the raster cartography, since it will be

used as the base positions for the survey and also because the available data are a lot and have to

74

4. Use case: verification and update of datasets needed for the evaluation of the environmental impact of theproduction of hydro-electricity

be previously harmonized.

The available data that could be used as support were (in shapefile format):

• the stream network

• the villages

• the municipalities

Instead the datasets that had to be integrated and validated were (in shapefile format):

• the offtakes

• the artificial network

• the dams

• the hydroelectric power plants

The work done on the vector data was to compare the shapefiles with the original project plans

in order to identify critical situations, in which incongruences where scoped and marked as to be

investigated.

4.3 Preparation of the tablet pc

Probably the most important thing to check before going out in the field is that the installed

software is properly working on the used hardware. For example the bluetooth connection between

the gps and the operation system is often a weak point and can lead to incredible loss of time when

already outdoors, which makes the fixing of the problems harder.

Another problems that was experienced is the breaking of the joins of the bag holding the tablet

pc tied around the neck. This is an issue that can make the survey a nightmare since it gets difficult

both to digitalize without sitting down, but also for example the carrying of the table pc in difficult

slope situations.

An important thing to consider is the battery time as opposed to the survey time. Usually a

tablet pc battery holds for half a day, so if the survey is planned to take the whole day, it will be

important either to have a second battery at hand, or the possibility to recharge the battery, as for

example during the lunch time at a restaurant.

The wrong preparation of the hardware used for the survey can transform the wanted successful

outcome of a survey into a frustrating fight with modern technologies in the open field. Don’t take

the risk, since technology in that case wins.

75

4. Use case: verification and update of datasets needed for the evaluation of the environmental impact of theproduction of hydro-electricity

4.4 Field mapping

Since we didn’t know were most of the smaller works were located we went out to the survey

with a technician of the public administration. Soon we found out that the same technician knew

perfectly the state of the art of the works, while he would not understand how to get started with

the digital maps.

The key point got to be able to translate the information on the map to real world locations to

which the guide could take us to investigate. After the first 20 minutes of incredulous grins, the

technician started to enjoy the new tools.

4.4.1 Use of the GPS and geonotes

The GPS is extremely useful in this kind of surveys first of all to show your actual position on

the base maps. This already helps to get started giving directions and finding the prior defined

points of interest.

Once you located the area you wanted to investigate, you can start to take a note of you position

and to verify the position of the artifact in question.

Very often it happens that since that data were digitalized without surveys, the position of the

artificial network can be right, whereas the work was placed wrong. For example image 5.2 shows

a case in which the point layer, which is the original data, contains wrong information. In the

particular case the offtake was placed on the artificial channel instead of upstream on the river

with a connection to the artificial channel.

Were possible the gps in the automatic mode was used to trace the paths along the artificial

channels. These informations were then taken to reconstruct the whole artificial network in those

cases in which the network was misaligned with the real situation. Figure 5.3 shows the case in

which the network was wrong in the original dataset.

In places were the data were completely missing (see 5.4 for an example), following the instruc-

tions of the technician of the local administration, the data were integrated through the use of the

gps and the digitalization tools. That way feature points were registered to identify offtakes and

stream nodes were sketched onto the map with the annotation tool in order to allow the user to

reconstruct the real path once back at the office.

At about four in the afternoon the tablet batteries were discharged. Still it has been possible to

finish the survey using the GeoPaparazzi application on the phone device.

It is really important to be able to quantify more or less the time a survey will take. Having

only one battery and having had only half an hour to recharge it during lunch we knew that the

76

4. Use case: verification and update of datasets needed for the evaluation of the environmental impact of theproduction of hydro-electricity

Figure 4.2: The blue pin showing the real position of the offtake, while in the original data the offtake is representedby the point and was placed on the intersection between the artificial channel and the river. The offtake will becorrected using the collected informations during the survey (see figure 5.5).

Figure 4.3: Example situation in which the artificial network in the original dataset was digitalized wrong. Thenetwork will be corrected using the collected informations during the survey (see figure 5.6).

77

4. Use case: verification and update of datasets needed for the evaluation of the environmental impact of theproduction of hydro-electricity

Figure 4.4: Example situation in which the artificial works are completely missing and have to be inserted based onthe collected information (see figure 5.7).

battery would not have last until the end of the survey. Luckily we were able to leave to the end of

the day those cases in which for just a point had to be collected to define the position of a work.

Those situation did not require the tablet pc with BeeGIS and we were able to survey them with

Geopaparazzi, which gives the possibility to take georeferenced note and pictures.

4.5 Back in the office

Once returned to the office the surveyed were used to review the situation of the original datasets

to correct and integrate them.

The first action we took was to export the gps log that had been traced by BeeGIS behind the

scenes and that is really important because it bounds the spatial information of the survey to its

temporal information, supplying a survey history (the export is described in section 3.6.1).

Since several pictures were taken during the survey to support the decision making, those were

imported into BeeGIS and synchronized with the gps log (see section 3.6.2).

Also the Geopaparazzi data have been imported to exploit the double gps log and the oriented

pictures, which make it very easy to quickly recall the real displacement of the works on the map

(see chapter 4.4).

Through all these information it has been possible to correct critical data errors present in the

78

4. Use case: verification and update of datasets needed for the evaluation of the environmental impact of theproduction of hydro-electricity

original data. A few examples are shown to give the reader a better idea about the work done:

Figure 5.5 shows the result of the correction of the position of the wrong placed offtake of figure

5.2. In this case the offtake, which had been placed in the wrong position, has been moved to its

real position and connected to the main channel through a connection pipe.

Figure 4.5: Comparison between the wrong (in the smaller frame on the right) and the corrected situation. Thewrong placed offtake has been replaced.

Figure 5.6 shows the result of the correction of wrong digitalized network part and offtakes of

figure 5.3. In this case the network channel has been digitalized wrong following the assumption

that it would pass through the offtake Rio Masere. Instead the survey showed, that the network had

a different direction and that the offtake was connected to the network through a connection pipe.

In this case not only geonotes were used to identify reference places as for example the intersection

of the road with the connection pipe, also the gps log was used to walk the real network position

to be able to better digitalize it once back in the office.

Figure 5.7 shows the result of the integration of missing data as of figure 5.4. This case has been

one of the worst, since almost all of the offtakes and many channel parts were missing, so that it

had been necessary to use a lot of geonotes to describe the real situation. Through the use of the

gps placed geonotes and the gps log all the offtakes, channel parts and connection pipes have been

created and integrated in the original dataset.

The corrected dataset was then used for an environmental analysis in which the amount of

each artificial channel per municipality had to be calculated. The result of the analysis would

influence an administrative decision workflow bound to sensible economic outcome, so the user can

understand why it had been so important to make sure that the artificial channels were digitalized

79

4. Use case: verification and update of datasets needed for the evaluation of the environmental impact of theproduction of hydro-electricity

Figure 4.6: Comparison between the wrong (in the smaller frame on the right) and the corrected situation.The wrongdigitalized channel and the wrong placed offtake have been replaced.

Figure 4.7: Comparison between the wrong (in the smaller frame on the right) and the corrected situation.Themissing data have been created and integrated.

80

4. Use case: verification and update of datasets needed for the evaluation of the environmental impact of theproduction of hydro-electricity

in the proper location and in the right length.

81

BIBLIOGRAPHY

Bibliography

[De Donatis et al., 2006] Mauro De Donatis, Lorenzo Bruciatelli, MAP IT: The GIS software for

field mapping with tablet pc, Computers & Geosciences 32 (2006) 673-680.

[Jones et al., 2004] Richard R. Jones, Kenneth J.W.McCaffrey, Robert W.Wilson, Robert

E.Holdsworth, Digital field data acquisition: towards increased quantification of uncertainty dur-

ing geological mapping., Geological Society, London, Special Publications, 239, 43-56

[De Donatis et al., 2005] Mauro De Donatis, Lorenzo Bruciatelli, Sara Susini, MAP IT: a GIS/GPS

software solution for digital field mapping, Digital Mapping Techniques 2005

[Clegg, De Donatis et al., 2006] P. Clegga, L. Bruciatelli, F. Domingosa, R.R. Jonesc, M. De Do-

natis, R.W. Wilsona Digital geological mapping with tablet PC and PDA: A comparison, Com-

puters & Geosciences 32 (2006) 1682–1698

[De Donatis] M. De Donatis, Earth and Environmental Sciences: field classes with GIS/GPS and

tablet PC

[Todd Williams et al., 2004] Eclipse-based Applications:Java on the Desktop Revisited, EclipseCon

2004 (http://www.eclipsecon.org/2004/EclipseCon 2004 TechnicalTrackPresentations/

17 Williams May Dovich.pdf).

[Steve Brody, 2001] Interview: The Eclipse code donation, DeveloperWorks Open source projects

zone editor, IBM (http://www.ibm.com/developerworks/linux/library/l-erick.html).

[Digital divide] (http://www.digitaldivide.net/articles/view.php?ArticleID=460)

[Bauer et al., 2007] Christian Bauer, Gavin King, Java Persistence with Hibernate, Manning Pub-

blications, 2007

[McKenzie, 2007] Cameron McKenzie, Hibernate made easy, McKenzie, 2007

[Ableson et al., 2007] Frank Ableson, Charlie Collins, Robi Sen, Unlocking Android, a developer’s

guide, Manning Pubblications, 2009

[Clayberg et al., 2009] Eric Clayberg, Dan Rubel, Eclipse Plug-ins, third edition, Addison Wesley,

2009

[McAffer et al. 2007] Jeff McAffer, Jean-Michel Lemieux ,Eclipse Rich Client Platform: Designing,

Coding, and Packaging Java(TM) Applications, Addison Wesley, 2007

82