The use of " image" in Geographical Information
Market :
results of an inquiry on the needs of end-users in urban studies.
Anne
Puissant1 & Christiane Weber2
Laboratoire Image et
Ville, ULP Strasbourg, 3 rue de l’Argonne , F- 67 000 Strasbourg
Tel : (33) 3 90 24 09 74 – Fax : (33) 3 90 24 09 50
anne@lorraine.u-strasbg.fr
and chris@lorraine.u-strasbg.fr
1
PhD Thesis
2
Researcher
1. A report :
the Geographical Information market, a sector in full expansion ?
Geographical
Information (GI) which was until the beginning of the Eighties a primarily
administrative activity is becoming today an economic sector in full evolution.
An information is called "geographical" if it refers to the
description of objects, events, phenomena located on the earth’s surface. If a
spatial localization is associated with information of different nature, these
become potentially geographical. The field of GI is thus very broad.
The development of this sector
became a political and economic stake due to the development of the Information
Society in the developed countries and in France as underlined by the PAGSI
(Plan d’Action Gouvernemental de la Société de l’Information, in 1999), by the
various reports carried out for the government (Report Mandelkern, in 1998, on
the "Information Society", and Report Lengagne in the early 1999, on
"the role of IGN in the GI sector") [Véler, 2000 ]. However, if this
development is general in the developed countries, the comparison of the growth
rate per country shows substantial differences from 10% to more than 30% per
year (Appendix 1). The analysis of the causes of these differences in Europe
shows that if in a lot of countries there are a leading public actor, the
positioning of this actor and the nature of its offer strongly influences the
dynamism of the activity [Lengagne, 1999].
Indeed, since the first launching of
earth observation satellites and with the development of the first numerical
cartography tools in the seventies, the GI actors have moved in an environment
subjected to fast changes. This sector in full numeric evolution currently
knows a triple technological, data processing and organizational revolution
[Ecobichon, 1994].
1.1 A technological
revolution
Satellite positioning system enables to obtain (in real or slightly differed time) a very precise measure of location with simple and cheap equipment (a few hundred euros). This technology offers anyone the possibility of measuring its position with accuracy and also modifies the methods to collect information. In addition recent and announced launchings of earth observation satellites with one meter or below one meter spatial resolution also promise to change the Information Society. In this context the use of digital images enables to facilitate automatic extraction methods and to reproduce a realistic view of earth without any degradation.
1.2 A data-processing
revolution
The data-processing development also
brought an accelerated automation and made it possible to multiply the
applications. The techniques of digitalisation, normalisation, storage and
management improved increasingly and performance being equal, considerably
reduced their costs. Data processing also opened this sector of activity to the
new professionals, in particular to the data processing specialists, sometimes
not quite conscious of the technical and conceptual complexities of GI. The
geographical dimension in the management and the analysis of information, which
until recent years required very specialised tools, is today better and better
integrated in Planning Agencies information system. Therefore the setting-up
and the management of large databases describing territory has become easier.
Thus, this new field of application of GI is approached in a majority of
planning services or urban management services through Geographical Information
Systems (GIS) capacities.
The influence of Internet and on line technologies on the functioning offer market and on the relations between the providers and the users of GI is still far from being completely evaluated. By allowing a direct relation between the source of information and the consumer, Internet is completely modifying the practices and the stakes.
1.3 An Organizational
revolution
These main evolutions, which have occurred in a context of institutional changes (European construction, opening of the public markets) and the consecutive austerity of the public budgets, disturbed the market [Ecobichon, 1994]. For instance due to the decentralization laws, in France, the different levels of local communities (Cities and Towns, Departments and Regions) play a major role in the management of their territory. They have invested considerably in the creation of GIS for decision-making in urban planning and management [Pornon, 1995].
2. Objectives
At the same time new communication
and information technologies allowing the multiplication
of available information on the market (offer) create a great
potential public and a profusion in which it is easy to be lost. The users have
to choose the required information carefully according to their goals and
tasks. Moreover, each user integrates and locates his own information and
becomes himself a producer of data prone to be spread and exchanged.
All these evolutions radically
modify the working methods for the producers and for the users of GI because
they induce new requirements, transform the nature of the relations between the
actors and change the stakes.
Several questions have already been
asked by some authors [Weber, 1991 ; Bailly, 1995 ; Roche, 1997]. Weber (1991)
emphasizes the questions on the effects of GIS on the spatial dynamics. Indeed,
it is significant to think about the new functions and the changes in
behaviours (evolution in the relations between producers and users, internal
practises changes in the organizations) due to the use of GIS in their
"spatial activities". Images (air photographs or satellite images)
are part of these new data sources available by the users. They are used by
urban managers through GIS. An analysis on the role of the images as potential
data source, and their impacts on the practices of the professionals have to be
conducted. However, the information about the use of these data and the needs
associated in GIS are uncommon or non-existent in the literature. In this
context, a survey was conducted in France and Europe, with the following
objectives :
Ø
to assess the role and the place of
images (air photographs and satellite images) in the practices of the users ;
Ø
to identify the users’ needs in
terms of image data adapted to their applications ;
Ø
to identify urban objects and
characteristics useful for the applications in urban areas.
This article
is part of a wider research (PhD Thesis), in the
continuity of the spatial program of the CNES (Centre National des Etudes
Spatiales). With the preparation of the program PLEIADES[1]
a specification of the different features is necessary, in particular the
"high optical resolution" component : indeed, which users must
this component satisfy, in particular in urban areas, and for what kind of
applications ; what are the characteristics of the information required, for
the observations (spatial, spectral and/or radiometric and geometrical
requirements) and for the services (programming, storage, time of production) ?
This article is more particularly directed towards the only questions about the activities of the organizations in GIS (period of activity in GIS, software used, acquisition and production of data), about the potential use of images in the decision-making and management processes and about the users’ needs in urban objects. After reminding of the inquired techniques used to collect information, we will present the treatments and their results. The investigation was made between June 1999 and 2000 with the professionals of urban planning and management (City Councils, Planning Agencies) in France and in a few European countries.
The interest of this investigation lies in a multi-user sample in order to contact the whole organizations working on urban areas. This approach is not common in the literature and the existing ones generally consist of inventories of phenomena and their evolutions and/or concern only one type of organization. The mid-term objective is to define standard behaviours of image users.
3. Methodology
3.1 The concerned
population
The population concerned in France and Europe contacted by our investigation is formed by all the actors working on urban areas : local authorities (City Councils, County), inter-city organizations (Urban Communities), Administrations and their services (for example, in France, the Center d’Etudes sur les Réseaux, les Transport et l’Urbanisme, the Direction Départementale de l’Equipement), associative structures (Planning Agencies) and independent structures such as network management agencies, surveyor offices, consultancy companies…
3.2 The data survey :
three consecutive stages
3.2.1
The questionnaire
3.2.1.1 The sampling
In order to maximize the users
contacted, the inquiry was diffused :
-
to some organizations, through inventories edited by the Consulting Company
IETI, and the French Ministry of Equipment, Transport and Housing ;
- to some
organizations less represented (for instance, 12 Urban Communities, 7 Centre
d’Etudes Techniques de l’Equipement), the questionnaire was diffused to the
total population[2].
3.2.1.2
The thematic
The
questionnaire is articulated around the following topics : the characteristics
of the organizations structures, the
activities of the organizations in localised data (GIS), the applications and
the useful objects, the input data and the use of the satellite images.
"Closed" questions (limiting the choice of the answers) were used to
test an assumption and "opened" questions when the objective was to
understand a phenomenon [De Singly, 1992].
3.2.1.3
The pre-tests
At first the questionnaire was pre-tested in order to collect comments on the contents, its comprehension, as well as qualitative information (shape and presentation). This pre-test was carried out, in May 1999, with 10 organizations selected for their geographical proximity of Strasbourg and for which a fast return could be hoped. This stage allowed us to change the formulation of the questions and the specific vocabulary used in the urban planning and management.
3.2.1.4
The final questionnaire
The final
version of the questionnaire was sent between June and October 1999 by post and
e-mail to 200 French organizations, and 150 European organizations (23 European
countries extended to the eastern countries). These figures[3]
do not take into account the questionnaire sent to newsgroups in GI (Georezo,
Géomatique, Cassini, Infogeo). A total of 33 questionnaires were completed (25
" French " and 8 " Europeans ") for an answer rate of 9
%. In comparison, an investigation carried out by the company SCOT
reached 6% of answers [Galaup et al., 1997].
The main difficulty was to contact the competent services and the person capable of answering, in addition to the unanswered questionnaires. Moreover, a few questionnaires were "bypassed" because some of them intended for planner and manager were directly transmitted to the service in charge of the data management. This phenomenon was neither expected, nor wished at the beginning.
3.2.2
A workshop to define the needs of end-users
The second
stage, consecutive to the actions launched by the French government (report
Lengagne in September 1999) was to gather together GI professionals in a
workshop (January 30 2000) entitled "Which scales
for which needs : utility of large scale GI for urban studies".
It was held in Strasbourg. The objective was to confront the individual
practices of the users (data used, problems) in order to highlight requirements
(in complement of the questionnaire) and to check the relevance of the
available data on the market (advantages and limits). Following the
presentation of the preliminary results of the survey, the reactions and the
discussions, the investigation was continued and reoriented.
3.2.3
Interviews to complete end-users needs
From February to October 2000, 25 interviews were carried out with local communities (Urban Communities and City Councils in France and in the neighbours countries), having acquired an orthophotograph[4]. Indeed, the workshop and the questionnaire emphasize the importance of the local communities as data producers and large scale GI users (particularly users of images data). These interviews also allowed to obtain a percentage of sufficient answers for quantitative processing. They were carried out from a grid with opened and closed questions about the two following topics : images data (orthophotographs and satellite images) and produced data (created, managed in the service).
Thus, by
combining the different inquiry techniques, the rate
of answers reaches 15 %. The people interviewed are in charge of
urban studies (30 %) and responsible for the data management (70 %). This
distribution is due partly to the bias coming from the transfer of the paper
questionnaires to persons in charge of the GIS and the actors privileged during
the interviews. This distribution has to be taken into account for the
interpretation of the results.
3.3 The data treatment
The
collected data were captured in a data base. A coding of the replies was realised
for the closed questions. The opened questions were analysed with a content analysis. This method consists in
building a reading grid and these translating the answers a posteriori in
the form of closed questions [Moscarola, 1990]. The answers (coded) were
exploited statistically by exploring, structuring
methodology (frequency table, cross tabulation, simple factorial
analysis of correspondence, hierarchical classification). The processing and
the tests were realised with the software MODALISA (version 4.0). The
unanswered questions were not taken into account during the quantitative
processing.
4. Results of the survey
Before the analysis and the interpretation of the results relating to the GIS and the use of images, the characteristics of the organizations investigated and their equipment in GIS will be specified, in order to understand the framework of the image data.
4.1
Answered organism
The
investigated organizations (Appendix 2) are divided into seven categories to
facilitate the reading of the results. The "Other" category regroups
private managers of networks (for instance, The Telecom). Communal
organizations ("cities" in the caption) weren’t grouped in order to
check if the size of the cities enables to differentiate the organizations. The
replies (Figure 1) come from more than 60 % of local authorities (Cities, Urban communities and
Regional and General Council), then Public Administration Services and finally
mixed organizations (Association) and private companies (Other).
Figure 1 : Replies by organization.
The geographical distribution of the
organizations (Figure 2) indicates a representation of the whole French territory. At the European level, seven
Europeans countries answered the questionnaire (20 % of the total reply).
The low answer rate in these countries does not make it possible to process the
data separately. The following results were obtained by taking these two groups
into account simultaneously.
Figure
2 : Replies in accordance with their geographical origin.
4.2 Equipment in GIS
In 2000, all the investigated
organizations are equipped or in process of being equipped with GIS except two
organizations.
4.2.1 Period of activity in GIS
Among those which possess a GIS (Figure 3), 47 % of the organizations have acquired a GIS for more than 5 years, 43 % have acquired it in the last five years, and 9 % have a GIS for less than one year. The organizations equipped with GIS for more than 5 years are, for more than a half, Urban Communities and Cities of more than 100,000 inhabitants. Those equipped for less than 5 years are as a rule General and Regional Councils and Public Administration Services. The organizations of the "Association" type (Planning Agencies) are the last to be equipped with GIS.
Figure 3 : Period of activity in GIS by
organisms (in percentage).
4.2.2 Software used
GIS softwares
used (an organization can have more than one software), is for 40 % MapInfo, for 30 % Arc view, for more than 10 % Geocity, and for less than 10 % Geoconcept, STAR and APIC (Figure 4). The totality of the answered
Public Administration Services use MapInfo software and a few are equipped with
complementary software, such as Arc view. This tendency is also emphasized in
the GIS inventory of the METL (1999), whose investigation shows that GIS are
not only represented by MapInfo (choice related to the policy of diffusion of
the METL, in France) but that the tasks of the organizations require
complementary tools such as AutoCAD for the drawing or Access for the
management of the data. The primarily Arc Info and Arc view choice by the organizations in charge of urban studies is
due to a wish to use GIS with all its functions (management, analyses,
simulation). The workstations softwares such as Geocity,
APIC and STAR (software precursors in this
field) are primarily used by the urban communities and cities equipped with GIS
for a long time (for instance, Strasbourg, Mulhouse, Lyon). Finally, the
majority of urban communities and cities now use Arc view and MapInfo.
Figure
4 :Software used by organisms (in percentage).
A structuring analysis of the data enables to differentiate three user groups according to their type of organizations, their period of activity and the software used (Appendix 3) :
ü
a first group
(group 1) with the organizations precursory in GIS : cities of more 100,000 inhabitants
and Urban communities with a GIS for more than 5 years, using Geocity Apic,
Star and Arc view software ;
ü
a second group
(group 2) with the Public Administration Service, and Regional, General Council, using mainly Mapinfo, with a GIS
for less than 5 years ;
ü
a third group
(group 3) with the Planning Agencies using a GIS for less than one year, with
Geoconcept and the cities of more than 50,000 inhabitants, using the software
STAR (for instance, Besançon, and Belgian organizations).
Therefore,
the identified tendency of our sample is
representative of the general tendency of GIS diffusion on the French territory
[Roche, 1997 ; IETI, 1998 and 2000]. Figure 5 summarizes this
evolution. In short, if in the early Eighties the Cities of more than 100,000
inhabitants and Urban Communities were the first to be equipped, the General
and Regional Council and the Public Administration Service have been largely
equipped during these last five years (since 1995). Only the CETE have
developed their GIS for a longer time (beginning of the Nineties). The Planning
Agencies started to be equipped for these last two years. For instance, the GIS
pole of Strasbourg Planning Agencies was created in 1999, the one in Belfort
2000.
This tendency is confirmed in Europe, except in the United Kingdom, Germany or the Netherlands, which are largely in advance compared to France. In fact, these countries have a dominating public actor. The position of this actor in these countries and the nature of its offer strongly influence the dynamism of the GI sector activity [Livre Blanc, 1998 ; Lengagne, 1999]. The low answer rate in the European countries does not enable to highlight the advance in these countries.
Figure
5 : Diffusion of GIS according to the time, and the type of organizations
in France [Puissant, 2000].
4.3
Activity in GIS
In the following paragraphs the analysis of the results has been conducted on the basis of working assumptions, which may constitute anticipated answers to the questions raised in § 2.
The analysis
of the GIS use by the actors has already been carried out in several
publications [Pornon, 1993, 1995 ; Roche, 1997]. The objective of the survey is
to check if the observed tendency in the literature is confirmed at present.
Indeed, the GIS would remain a tool of
technicians and in a lesser degree a tool for thematicians, often
beyond of reach non-specialists (in particular decision-makers). The capacities
of the current systems would generally be adapted to technical activities in
order to provide information at an operational
level [Laaribi et al., 1993]. Often they are only used to store data in layers and to cartography data with spatial reference in
order to facilitate the operational management of the territory. The assumption
to check is as follows :
Working
Hypothesis n°1 :
At present
GIS are only used as a simple managing and structuring tool.
Before determining the levels of the GIS use, it is necessary to know the GI applications’ levels existing in urban planning and urban management. In the literature, four categories of GI users (Table 1) can be identified according to their applications, the data scales used in input and the types of documents created in output (output) [Reports, 2000 ; MERCATOR, 1999 ; Eckmann, 1997]. The first three groups correspond to the three levels of responsibilities defined by Chevalier (1990). Figure 6 shows each level of this pyramid, anyone corresponding to a particular actor with a level of responsibility, associated to a requirement in data sources, which is characterized by the volume of necessary data, and the level of generalisation, precision and form under which this information must be given to the interested parties.
The decisions
(strategic level) concern the decision-makers,
the planning tasks (tactical level) concern the directions of the
services (in the case of a local territory or a Planning Agency, for instance),
and the practical execution of the
management of the territory (operational level)
concern the technical professionals.
|
Table 1 :
Summaries of the user categories and their requirements |
|||
|
User
category |
Applications
|
Data scales
(input) |
Output
document (output) |
|
Manager of the
territory (G1) |
Technical
management, basic
mapping, (network, urban green space management …) |
1:200 to
1:500 1:1000 to
1:2000 (maximum 1:10000) |
Technical
network plans, Topographic plan, … |
|
Analyst of
the territory (G2) |
Urban planning,
operational planning, impact study, social, urban inventory and evolutions,
.. |
1:5000 to
1:10000 |
Planning
document (POS, …) and thematic cartography,
... |
|
Answers of
synthesis (G3) |
Distribution analysis of phenomenon, objects, prospective,
... |
1:10000 to
1:1000000 |
Thematic,
synthetic cartography, management simulation, … |
|
Data
manager (G4) |
1:200 to
1:1000000 |
/ |
|
This horizontal diagram of decision (Figure 6), where the relations between the levels of decision appear, can be completed by a "vertical" level : the "data management” (structuring, formatting, controlling, exchanging). This vertical structure makes the connection between each horizontal level, by ensuring the exchange of the data irrespective of the scales and their levels of application. The results of the inquiry confirm the relevance of the assumption according to which GIS are still for most people a data management tool. However, our results highlight that GIS are more and more used, in most of the organizations, for concrete applications, primarily for their functions of cartography, as communication tool and as a decision-making help. The GIS field of applications is more diversified and is growing (multi-task SIG).
Figure 6 : The decision pyramid of actors and
their requirements
[Puissant,
2000 : inspired by Chevalier,
1990].
Nevertheless,
GIS have not got become analysis tools for most users (except in some cases
for Urban Communities) because these functions are sometimes more difficult to
set up. The first constraint to the GIS use are the constraints
of the data (quality, in terms of scales, accuracy, exhaustively,
acquisition problem, exchange, unavailability). Then, the technical (interface and availability of
application), social (lack of
education, conflicts of power) and finally economic
constraints emphasize the slow
diffusion of GIS in the organizations [Interviews 2000, Sureau et al., 2001].
The predominance of the MapInfo software (§ 4.2) also highlights that GIS are
not used for their analysis functions but especially for their cartographic and
visualization functions. In addition, it is often difficult to convince the
users that their methods (practical) "are sometimes old-fashioned"
and could be improved by the use of this tool (with the automation of some
treatments, for instance). Additional constraints related to the lack of implication, awareness, and knowledge
of the users, do not favour the use of GIS.
4.4 The uses of
spatial data
In this
paragraph, the objective is to define the image
use level (air photographs, orthophotographs, satellite images) in
GIS. So, one must be determine if a link exists between the potential quality
of GI of image type (according to the characteristics of the available products
: spatial, spectral, temporal resolution) and its use in the practices of urban
management. An anticipated answer to this question could be as the following :
Working
Hypothesis n°2 :
The professionals
use image data in a GIS as a simple communication support.
The analysis and the interpretation of the results on the use of the image in urban areas must be qualified, according to the utilisation of an orthophotograph or of a satellite image. Indeed, whereas the orthophotograph is used by more than half the surveyed organizations (this tendency increases more and more), the satellite image is only used by approximately a third of the organizations.
4.4.1 The uses of ORTHOPHOTOGRAPHS
The
results of the survey (Table 2) show an increasing use of this document in GIS,
irrespective of the organizations, first as communication
support, enabling to present to the public an objective and
realistic vision of the territory, easily readable by the decision-makers and
the local population who "recognize the territory".
Then,
the orthophotograph is especially used as a large scale source of information for the management of the territory,
and more precisely :
ü
for the creation of reference data : creation of
information layers, in a city without any other data ;
ü
for the management of these data : update of
existing information layers, in city already having reference data.
|
Table 2 : Use
of orthophotography |
||
|
Information source |
Communication support |
System of
reference |
|
54 % |
33 % |
13 % |
4.4.2
The uses of SATELLITE IMAGES
The results (Table 3) show that the current satellite images are not used or little in urban areas as a communication support. Indeed, these images and their characteristics, in particular in terms of spectral resolution (commonly called by the users the "color") and of spatial resolution (compared with the grain of air photography), are not appropriate to the users, because this "false colors" and "microscopic" view is not easily and directly interpretable by the public and the decision makers. On the other hand, the surface covered by an image (scene) seems a decision criterion by the people in charge to this type of data.
|
Table 3 : Use of satellite
images |
||||
|
Land cover
cartography |
Identification
of green areas |
Identification
of artificialized surfaces |
Identification
of changes |
Complementary
view Communication
support |
|
50 % |
22 % |
11 % |
11 % |
6 % |
The
acquisition of an orthophotograph on an agglomeration does not inevitably slow
down the decision to acquire a satellite image : for instance the city of Le Havre
(France) possesses an orthophotograph and several satellite images (SPOT P and
XS), and the Urban Community of Strasbourg (orthophotograph of 20 cm)
intends to acquire a satellite image soon [Interviews, 2000]. The crossed
analysis of the replies shows that satellite image is mostly used for the
production of land cover type cartography,
on a regional scale (1:25000), mainly
by the Regional and General Councils and the Planning Agencies. Then, it is
used as an information source for the constitution of specific information layers (green areas,
parks, impermeable surfaces, detection of the change), on large scales
(1:10000), and with a lesser precision than the orthophotographs (centimetre
accuracy for the orthophotographs, one meter accuracy for the satellite
images), primarily by the Urban Communities and the cities of more than 100,000
inhabitants.
These two
facts allow us to check the relevance of assumption two and to assume that if
an orthophotograph is initially used
as a communication support, it is not the case of the satellite image. On the
other hand, if the second use of the orthophotograph is to be an information source on a large scale (1:200
to the 1:2000), this one
constitutes the main use of satellite image,
in urban areas, but on a smaller scale
(1:25000). The explanations of these uses are mainly linked to the spatial and
spectral resolution of these products.
Thus, these two types of data are not competing but rather complementary, in terms of applications, and of production scales. The current characteristics of spatial and spectral resolution can also explain the non-use of the satellite image as communication support and information source by the professionals of the urban areas. This assumption can be formulated as follows :
Working
Hypothesis n°3 :
The non-use
of satellite images in GIS is due to the inadequacy of their spatial
resolution.
The combined analysis of the problems met with (Table 4) and of the explanation of the non-use of current images (Table 5) indicates that the spatial resolution is one of the brakes and one of the main reasons for the non-use of satellite images.
|
Table
4 : Problems encountered in the use of the satellite images (on 16 % of
responses). |
||||
|
Pixel size |
Difficulty
of automation |
Incompatibility
of nomenclature |
Difficulty
of interpretation |
Complexity
of software |
|
25 % |
25 % |
25 % |
13 % |
12 % |
|
Table
5 : Explications
of the non-use satellite images |
|||||
|
Pixel size |
Cost and
lack of financing |
Satisfaction |
Lack of knowledge |
Technical
difficulties (interpretation and software) |
Delivery
time |
|
40 % |
24 % |
20 % |
10 % |
4 % |
2 % |
On the one hand, the users who
already possess very high resolution (below one meter) orthophotographs, have
difficulties in working with lower resolution data. They have the impression of
"viewing nothing". On the other hand, for the users who know
these data, the low spatial resolution of the current images (except Ikonos) is
a problem for the realization of their applications, especially in dense urban
areas. Indeed preceding works [Scot, 1997; Puissant, 1998; Puissant and Weber,
1999 ] have demonstrated that only strategic level application, with scales
from 1:25000 to 1:1000000, can be carried out with satellite image (Table
6). But as we have underlined previously, the tactical level
applications represent, with the territory management, the main applications
for the users in urban areas.
|
Table 6 : Scales
reached by satellite images compared to urban applications |
||||
|
Scales of applications in urban
areas |
|
Scales
reaches by satellite images |
||
|
Applications |
Scales |
|
Satellite Images |
Scales |
|
Technical
Management |
1:200 to
1:500 |
Þ |
? |
|
|
Basic mapping |
1:1000 to
1:2000 |
Þ |
? |
|
|
Urban planning |
1:5000 to
1:10000 |
Þ |
? |
|
|
Abstract level |
1:10000 to
1:1000000 |
Þ |
SPOT P and XS |
1:25000 to 1:50000 |
|
|
|
|
Landsat TM |
1:100000 |
|
|
|
|
Landsat MSS |
> 1:100000 |
In addition
to the spatial resolution, the cost of the images
and the lack of financing of the organizations are the second constraint limiting the diffusion of
satellite images.
The lack
of knowledge on the potentialities of satellite images and even on
their existence is a third constraint.
The latter is linked to the satisfaction of the users to the different used
data which involve a non-awareness of the users towards these new products. In
addition, when the economic constraint is not present, the users must prove to
their hierarchy the interest and the contribution of this type of data compared
to the existing data. The awareness and the motivation of the people are thus a
significant factor directly governing the purchase of these data.
Then, technical
difficulties (software, interpretation of the data) induce the users
to sub-contract the satellite image processing and thus increase the costs, and
often extend the delivery time of the final product increases. These elements
slow down the users unwilling to change.
Finally, two
significant constraints limiting the use of the images, are the incompatibility of the nomenclatures obtained
from the images to the nomenclatures of the land cover, and the difficulty of automate the procedures of extraction. This problem can be explained by
the fact that the users need to extract the objects and their attributes
(surfaces, form) rather than zones.
In short, if
it is true that the spatial resolution is a constraint and the main reason of
the non-use of the images, others constraints, not unimportant, must be taken
into account. Figure 7 simplifies and summarizes the different
constraints to the use of satellite image, as an information source and a
communication tool.
Figure
7 : Constraints and explications of the non use of the satellite image
[Puissant, 2001].
4.5 The needs
of end-users in spatial data
The results
of the survey show, in agreement with the answers obtained on the use of the
satellite images, that all the organizations interviewed are waiting for value-added image products. The
first users of these products are the Regional and General Councils and the
Public Administration Services. For these organizations which work at on macro
scale (department or region), a one-meter resolution would be enough. The
second expectation of the users is linked to the spatial
resolution. The need of resolution below one meter is primarily
expressed by Urban Communities and Cities of all sizes. Compression facilities are required by
organizations of a smaller size with a limited budget (Public Administration
Services, few cities with less than 100,000 inhabitants). The need for coherent nomenclature with the land cover plan
is mainly stressed by Urban Communities (Table 7), this need is
finally expressed by few organizations because few of them know the
potentialities of the images.
|
Table
7 : Expectation of users in satellite images. |
||||
|
Value-added
product |
One meter
resolution |
< 1 m
resolution |
Facility of
compression |
Coherent
nomenclature |
|
47 % |
31 % |
13 % |
6 % |
3 % |
Others needs or expectations which the
non-users are not aware, can be deduced from the preceding analyses (problems
encountered by the users of the orthophotographs, of the satellite images and
factors has account for their non-use). It is the case for the dropped shadows, the uniformity of the radiometry,
the volume of the data and the delay to obtain the products (last problem only
mentioned for the users of the orthophotographs). Indeed, a few users have
already had the possibilities to visualize one meter images such as Ikonos.
However, the problem of the hidden areas will disturb the users in their task
since the future and expected image resolutions will approach the resolution of
the orthophotographs. The uniformity of radiometry is essential for the
communication but also for the automatic recognition. This need lends to a need
for single cover (not a mosaic) of the studied zone, with limits adapted to the
interest area (agglomeration…).
A multivariate analysis highlighted four "profiles of satellite image users" (Appendix 4) :
ü
the first group
(group 1) includes Urban Communities already using the image (Nancy, Brest and
Dunkerque) for the identification of the changes, the cartography of the green
areas and impermeable surfaces. These Urban Communities expect an resolution
below one meter in dense urban areas and nomenclature adapted to the POS ;
ü
the second group
(group 2) isolates Urban Communities which do not use images for the time
being, but where reflections are in progress (Strasbourg, Bordeaux and Lyon)
and whose expectations depend on the orthophotographs they have ;
ü
a third group
(group 3) is composed of the majority of the Planning Agencies, which use the
image for the cartography of land cover and to have a complementary view of
their territory. This group is confronted with problems involved in the
automation of the processing and interpretation ;
ü
the fourth group
(group 4) include organizations which do not lay out images (General and
Regional Council, Public Administration Service, cities between 50,000 and more
than 100,000 inhabitants). Their expectations concern one meter resolution
(lower than 5 m) and value-added products such as land cover cartography.
5. Conclusion
The information collected in the literature and through our investigation has smalled to understand the place of the image in the users applications and to determine their requirements. Currently, the significant facts to retain are :
- An increased tendency of a MULTI-TASK, and MULTI-USER use of the GIS, and sometimes a multi-software use (CAO/DAO software used
in complement of the GIS).
- The majority of the
applications in urban areas (related to GIS) concern the management of the
data (operational level), and the
urban planning (tactical level).
These applications use geographical data from the 1:200 to the 1:10000.
- The image has two precise
functions, it is used as a communication
tool and as an information source.
Orthophotograph initially plays only a role of communication support.
- Orthophotograph is a used
product more and more as an information source, on a large scale in
particular in the field of the operational management of the territory and
in default of the others products, degraded orthophotographs with lower
resolutions, are used in the field of urban planning and management.
5.
Satellite
image is especially used for its function of information source on regional
scales (1:25000), to cartography the land cover or specific objects such as
green areas (Table 8).
|
Table 8: Applications
and used images |
||||
|
Scales of applications in urban
areas |
|
Currently images data used to
answer to these applications |
||
|
Applications |
Scales |
|
Images |
Resolution |
|
Technical management |
1:200 to
1:500 |
Þ |
Orthophotograph |
20 cm |
|
Basic mapping |
1:1000 to
1:2000 |
Þ |
Orthophotograph |
20 to 50 cm |
|
Urban planning |
1:5000 to
1:10000 |
Þ |
Orthophotograph |
50 cm to 1 m |
|
Prospective |
1:10000 to
1:1000000 |
Þ |
SPOT P and XS / Landsat |
10 to 30 m |
Satellite image will only be able to play a role in urban tasks if it can be used at the same time as a communication support and as an information source, in the scales between the 1:2000 and the 1:10000, the more used by the users.
To play a
role as a communication support, the
satellite image will have :
ü
to set of natural colors, directly interpretable
by non-specific users, and uniform radiometry
ü
to reach accurate geometrical details at least
one meter resolution.
To be used
as an information source, the
satellite image will have :
ü
to be available as a value added product
allowing an updating at least annual, available “on hand”, according to a very
flexible limit, at reasonable cost ;
ü
to cover large territories in order to lay out
the most uniform possible radiometry and thus to overcome one of the problems
encountered on orthophotographs.
For a raw image use (with pre-treatment such as geometrical and radiometric corrections), the users will have to lay out :
ü
automatic extraction and compression procedures
of the images ;
ü
products where the problem of the shadows and
the hidden areas are limited ;
ü
objects nomenclature adapted at least to the
nomenclature of the land cover plan.
In this context, do the current
image products already meet all these needs ? What can be the role and the
place of the future one meter images in the applications of the users in urban
areas ? These questions form the subject of a search in a PhD thesis. More
precisely, the objective of this work are, first, define image user types and
their applications ; secondly, identify potential applications of the existing
and future very high spatial resolution images (one meter or below one meter
resolution) in urban areas and to define the adapted characteristics (spatial,
spectral resolution…) of a new sensor to these urban applications.
References
Bailly A.,
Introduction au débat: perspectives en géographie de l’information et de la
communication, Sciences de la société, Colloque territoire, Société et
Communication, PUM, Toulouse, n°35, pp. 15-19, 1995.
Chevalier J., Implementation of an SIRS in a
municipal environment : a problem of computarisation or a problem of
organization ?, in Actes du Colloque : La Géomatique, voir…à sa mesure,
Association de géomatique municipal du Québec, Montréal, 1990.
Eckmann O., Gestion classique des données
spatiales et besoins actuels : difficultés d'appréciation de la demande et
propositions de réponses, l'exemple du couplage Phocus-Autocad, Thèse de
doctorat, Strasbourg, 296 p., 1997.
Ecobichon C., L’information géographique,
Hermès, Paris, 104 p., 1994.
De Singly, L'enquête et ses méthodes : le
questionnaire, Nathan, 1992
Fellous J.L., Le programme Pléiades, in
"Atelier Besoins Pléiades", CNES, Toulouse, 2000
Galaup M, De
Boissezon H., 1997, User workshops to define the
requirements of town/city local governments departments, Joint Research
Center, CEO programme, Final Report, , vol. 1 et 2, 130 p.
Georget E., Annuaire
1999 des SIG du METL, Publication du Centre d’études sur les réseaux, les
transports, l’urbanisme et les constructions publiques, 126 p., 2000.
IETI Consultants, L’observatoire Géomatique :
mise à jour 1998 et 2000, www.ieti.fr/1sig_observatoire.htm, 1998 et
2000.
Laaribi A., Chevalier J.J.,
Martel J.M.,
"Méthodologie d’intégration des SIG et de l’analyse multicritère", Revue
Internationale de Géomatique, Hermès, Paris, vol. 3, n°4, pp. 415-435, 1993
Lengagne G., Les perspectives d’évolution de
l’information géographique et les conséquences pour l’IGN, Rapport présenté au
Premier Ministre, 57 p., Septembre 1999.
Livre Blanc, L'information géographique
française dans la société de l'information : État des lieux et
propositions d'action, Rapport du gouvernement, 28 p., 1998.
Livre Vert, L’information émanant du secteur
public dans la société de l’information, France, 50 p., 1998.
Mercator, Le guide de l’information
géographique, Tome 1 et 2, Euro vista, Paris, 1999.
Moscarola J., Enquête et analyse de données, Paris, Vuibert Gestion, 1990.
Pornon H., "Utilisation et place des SIG dans les systèmes d’information
des organisations", Revue internationale de Géomatique, Hermès, Paris, vol. 3, n° 1-2,
pp.65-70, 1993.
Pornon H., Les SIG : mise en œuvre
et applications, Hermès, Paris, 160 p., 1995.
Puissant A., Les potentialités des images à très haute résolution
spatiale (THRS) pour la reconnaissance des composantes urbaines, Mémoire de
DEA, UFR de Géographie, ULP Strasbourg, 110 p, 1998.
Puissant A., Weber C., Les potentialités des images Très Haute Résolution
pour la reconnaissance des composantes urbaines, Journée scientifique
"Télédétection à THR et analyse d'images", 14 décembre 1999,
Montpellier, in Actes de la journée , 12 p., 1999.
Roche S., Enjeux
de l’appropriation sociale des technologies de l’information géographique pour
l’aménagement territorial : Etude de cas en France et au Québec, Thèse
de doctorat, Université d’Angers, 362 p., 1997.
Sureau K., Roche S., Caron C., Enquête sur les utilisateurs
et les utilisations du SIG, Mémoire de Maîtrise, Université d’Angers, en
cours (2001).
Véler B., La politique française dans le
domaine de l’information géographique, Géorama.net : Magazine
d’information géographique, 3 p., Janvier 2000.
Véler B., La politique d’information
géographique en Europe, Géorama.net : Magazine d’information
géographique, 2 p., Janvier 2000.
Weber C.,
"Les systèmes d’information géographique : une mode ou un nouveau concept
pour l’aménagement de l’espace", Revue des sciences de l’information
géographique et de l’analyse spatiale, Hermès, Paris, vol.1, n°1, pp11-21,
1991.
Appendix
1 – THE GEOGRAPHIC INFORMATION MARKET IN THE WORLD.
[Sources : Livre
Blanc, 1998 ; Report Lengagne, 1999 ; Ressource Naturelle Canada,
1999]
In Europe,
the GI market or the economic activity, has given rise these last few years to many studies in
order to assess them and compare the national dynamic. These studies realised
by specialized companies, the European Commission or the National Council GI
(in France) lead to absolute results which are not directly exploitable because
they define the activity in a different way. The relative levels of the various
national markets remain however similar whatever the method evaluation adopted
and the estimates of growth are also very close. Thus, these are the relative
positions which must be examined knowing that the European market of direct
sale of data to the end-user (except software, hardware, provisions of services
and subcontracting) is evaluated between 3.6 and 8.4 billion francs.
The
result for the 5 countries concerned by the study led for the CNIG gives the
following table where the French market was conventionally affected value 100 :
Comparison of the GI activity
between European countries (1998).
|
Country |
Market
in absolute value |
Market
report on PIB |
Estimated
growing on 5 years |
|
France |
100 |
100 |
50% |
|
Sweden |
23 |
150 |
100% |
|
United
Kingdom |
162 |
217 |
80% |
|
Netherlands |
33 |
133 |
250% |
|
Spain |
23 |
58 |
150% |
|
Germany |
231 |
158 |
80% |
On a
world scale, reports have also been developed but these study the geomatic
market. This one grows at the rate of 20 % per year. Currently, its value ranges
between 10 and 20 milliards of dollars. The geomatic industry gathers the
disciplines like surveying, aerial photography, remote sensing, the GIS.
Nowadays, the main market is north America and Europe. New markets are
beginning to develop in Latin America, East Europe and Asia. The United States
are classified in first returns for the technological development, Canada comes
in second place, followed by France, Germany, and the United Kingdom.
Appendix 2 – LIST OF THE ANSWERED
ORGANIZATIONS
|
Answered organisms |
Category |
|
[Individu 1 ; Agence
d’Urbanisme de Lilles] |
Association |
|
[Individu 2 ; DIREN
Nord-pas-de-Calais] |
Public Administration Service |
|
[Individu 3 ; CG Yvelines] |
General and Regional Council |
|
[Individu 4 ; Communauté
Urbaine de Brest] |
Urban Community |
|
[Individu 5 ; DGE, Montpellier] |
General and Regional Council |
|
[Individu 6 ; Région
Pays de Loire] |
General and Regional Council |
|
[Individu 7 ; CG
Calvados] |
General and Regional Council |
|
[Individu 8 ; DIREN
Alsace] |
Public
Administration Service |
|
[Individu 9 ;
SERTIT] |
Other |
|
[Individu 10 ;
Agence d’Urbanisme de Metz] |
Association |
|
[Individu 11; Agence
d’Urbanisme de Brest] |
Association |
|
[Individu 12 ; DDE
32 ; Gers] |
Public
Administration Service |
|
[Individu 13 ; DDE
69 ; Rhône] |
Public
Administration Service |
|
[Individu 14 ; CETE
de l’Est] |
Public
Administration Service |
|
[Individu 15 ; CG
Val de Marne] |
General and Regional Council |
|
[Individu 16 ; DDE
81 ; Albi] |
Public
Administration Service |
|
[Individu 17 ; DDE
65 ; Tarbes] |
Public
Administration Service |
|
[Individu 18 ; DDE
67 ; Bas-Rhin] |
Public
Administration Service |
|
[Individu 19 ; CG 67
; Bas-Rhin] |
General and Regional Council |
|
[Individu 20 ; CG 21 ;
Bourgogne] |
General and Regional Council |
|
[Individu 21 ; Ville
de Genève] |
City of more than 100,000
inhabitants |
|
[Individu 22 ;
AUDIAR] |
Association |
|
[Individu 23 ;
IAURIF] |
Association |
|
[Individu 24 ;
France Telecom] |
Other |
|
[Individu 25 ;
Ville de Bourgas ; Bulgarie] |
City of more than 100,000
inhabitants |
|
[Individu 26 ; Ville
de Birmingham ; UK] |
City of more than 100,000
inhabitants |
|
[Individu 27 ;
Agence de l’Environnement ; Cheshire ; UK] |
Association |
|
[Individu 28 ;
DATU ; Barcelone, S] |
General and Regional Council |
|
[Individu 29 ; Ville
de Padoue ; I] |
City between 50,000 et 100,000
inhabitants |
|
[Individu 30 ;
Agence d’Urbanisme de Leipzig ; DE] |
Association |
|
[Individu 31 ; Direction
Régionale pour la circulation ; DE] |
General and Regional Council |
|
[Individu 32 ;
Service circulation de Dresde ; DE] |
City of more than 100,000
inhabitants |
|
[Individu 33 ;
Société TRANSFER] |
Other |
|
[Individu 34 ; Ville
de le Havre] |
City of more than 100,000
inhabitants |
|
[Individu 35 ; DDE
Caen] |
Public
Administration Service |
|
[Individu 36 ; Ville
de Mulhouse] |
City of more than 100,000
inhabitants |
|
[Individu 37 ; Ville
de Metz] |
City of more than 100,000
inhabitants |
|
[Individu 38 ; Ville
de Belfort] |
City between 50,000 et 100,000
inhabitants |
|
[Individu 39 ;
Communauté Urbaine de Strasbourg] |
Urban
Community |
|
[Individu 40 ;
Communauté Urbaine de Nancy] |
Urban
Community |
|
[Individu 41 ; Ville
de Grenoble] |
City of more than 100,000
inhabitants |
|
[Individu 42 ; Ville
de Chambéry] |
City between 50,000 et 100,000
inhabitants |
|
[Individu 43 ;
Communauté urbaine de Lyon] |
Urban
Community |
|
[Individu 44 ; Ville
de Besançon] |
City of more than 100,000 inhabitants |
|
[Individu 45 ; ville
de Toulouse] |
City of more than 100,000
inhabitants |
|
[Individu 46 ;
district de Rennes] |
General and Regional Council |
|
[Individu 47 ;
Communauté Urbaine de Bordeaux] |
Urban
Community |
|
[Individu 48 ; ville
de Rennes] |
City of more than 100,000
inhabitants |
|
[Individu 49 ;
AURAN] |
Association |
|
[Individu 50 ; ville
de Nantes + questionnaire papier] |
City of more than 100,000
inhabitants |
|
[Individu 51 ;
Communauté Urbaine de Lilles] |
Urban
Community |
|
[Individu 52 ; Ville
d'Annecy] |
City between 50,000 et 100,000
inhabitants |
|
[Individu 53 ;
Communauté Urbaine de Dunkerque] |
Urban
Community |
|
[Individu 54 ; MET,
Belgique] |
General and Regional Council |
|
[Individu 55 ; Ville
de Mons ; Belgique ] |
City between 50,000 et 100,000
inhabitants |
|
[Individu 56 ; Ville
de Poitiers] |
City between 50,000 et 100,000
inhabitants |
Appendix 3 – HIERACHICAL
CLASSIFICATION ON THE FACTOR 1 AND 2
Questions : "ORGANIZATION TYPES", "PERIOD OF ACTIVITY IN GIS" and "SOFTWARE
USED"
|
Factor 1 |
Factor 2 |
Factor 3 |
|
36 % |
11 % |
10 % |
Appendix 4 –
HIERACHICAL CLASSIFICATION ON THE FACTOR 1 AND 2
Questions : "ORGANIZATION TYPES",
"image use", "problems", and "expectation"
|
Factor 1 |
Factor 2 |
Factor 3 |
|
25 % |
14 % |
10 % |
