International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences, Vol XXXV, Part B4. Istanbul 2004
from these sources comprises the same large effort as well. On
a long term a trend is pointing towards a decentralised data
storage. Therefore implementation of international standards
and adaptation of data models from national geospatial data
bases are necessary and show the requirements of methods
dealing with integration of decentralised data 'on the fly'.
3. THE GIMODIG PROJECT
The objective of the GiMoDig project (Geospatial Info-
Mobility Service by Real-Time Data-Integration and
Generalisation) is to find solutions for these problems. Funded
by the European Union, GiMoDig started in November, 2001
with a duration of 3 years. The target is to develop and test
methods for delivering harmonised, European, large-scale
geospatial data to a mobile user by means of real-time data-
integration and generalisation. The project aims at the creation
of a seamless data service, providing access through a common
interface, to the primary topographic geodatabases maintained
by the National Mapping Agencies (NMAs) in various
countries. The Finnish Geodetic Institute acts as a coordinator
for the project. The other participants are the Institute of
Cartography and Geoinformatics at the University of Hanover,
Federal Agency for Cartography and Geodesy (Germany),
National Survey and Cadastre (Denmark), National Land
Survey of Sweden and National Land Survey of Finland.
For the GiMoDig project we use the national topographic data
delivered by the NMAs in vector format. Therefore e.g.
zooming in does not cause any quality loss in map visualisation.
The access to the data in the GiMoDig project is not
implemented as offline but as online application. According to
his current position the mobile user receives the appropriate
map on his mobile device — again on the condition that the
device is able to locate itself. No additional work has to be done
before hand e.g. for selecting of the relevant map piece and the
storing on the mobile device.
Finally, when the user is moving in a border area he receives
the national topographic data of both countries. Here arises the
real challenge — the harmonisation of the national data. They
are often very heterogeneous, a uniform visualisation is not
always possible offhand.
Therefore the focus in the GiMoDig-project also lies on the
harmonisation of core national topographic databases and will
be described in more detail in the following sections.
This paper has the emphasis on the conceptual aspects of the
work. The details about the implementation are given e.g.
(Lehto, 2003), which documents the system architecture of the
prototype system. A very relevant paper is also (Lehto and
Sarjakoski, 2004) describing how schema translations have
been implemented using XSLT (Extensible Stylesheet
Language Transformations). Other documents of the project can
be found at the project's website (GiMoDig, 2004).
4. HARMONISATION STEP BY STEP
Due to historical, organisational and technical constraints the
core topographic datasets of the National Mapping Agencies
(NMAs) in the GiMoDig Consortium — and elsewhere — are
built on different conceptual models. For usage of national data
in cross-border applications a common schema has to be
defined and the necessary mappings from the national models to
the common schema have to be identified. Following the
terminology used in database modelling, the common GiMoDig
schema is referred to as 'Global Schema’, while a national
schema is referred to as 'Local Schema' in this paper.
130
At the same time harmonisation operations have to be defined
in order to achieve an automatic transformation from the
National to the Global Schema. The Global Schema is to be
defined in a way that minimizes the efforts of data
harmonisation from national databases. The harmonisation
operations from Local Schema to Global Schema have to be
limited to those procedures that can be performed on-the-fly in
real time when the data in national data bases is accessed.
In GiMoDig the process for creating the Global Schema was
realized out in several phases.
4.1 Inventory of the national topographic databases
As a first step in the process, an inventory of the national
topographic databases for all map scales is done. A survey on
the availability of topographic vector data in Germany,
Denmark, Finland and Sweden reveals that all countries can
provide national data at resolutions 1:5.000 to 1:10.000,
1:250.000 and 1:1.000.000, whereas data at resolution 1:25.000
to 1:50.000 is available in Finland and Sweden only (Illert,
Afflerbach, 2003a).
4.2 Classification into the FACC-Code
The classification of the national topographic data of the
participating countries according to the FACC-Code (Feature
Attribute Coding Catalogue) is done using this inventory as the
background information. The FACC was originally created by
NATO as a military standard within the scope of DIGEST
(Digital Geographic Information Exchange Standard). It is
divided into object types, attribute types and attribute values.
Further on it contains notifications in different languages,
alphanumeric codes and class definitions. Created mainly for
military purposes the FACC is now also used for civil
applications. It is already used for other international projects of
the National Mapping Agencies. EuroGeographics (2003), the
organisation of European NMAs, prepares for pan-European
topographic ^ databases at a scale of 1:250.000
(EuroRegionalMap (2003)) and 1:1.000.000 (EuroGlobalMap
(2003)).
In terms of consistency across scales the GiMoDig Global
Schema will be compliant with the conceptual models of ERM
and EGM. This condition applies to the conceptual schema but
not to the data models, because the GiMoDig project aims at
on-line harmonisation in real time from distributed sources
while ERM and EGM constitute harmonised datasets that exist
physically. The production of ERM and EGM includes
interactive work, e.g. on edge matching, thus facilitating cross-
border topology which is not a feature in the GiMoDig project.
With the awareness of these pan-European projects at medium
and small scales, the efforts in GiMoDig are focused on the
base data at large scales (1.5.000 to 1:10.000), thus
complementing the other projects. In order to be compliant with
ERM and EGM, the specifications of the GiMoDig Global
Schema are based on the DIGEST FACC.
The following topographic data bases are used for the common
Global Schema:
Finland: Topographic Database
Denmark: TOPIODK
Germany: ATKIS Base-DLM
Sweden: the Geographic Sweden Data (GSD)
The Global Schema will provide an umbrella over these
databases. As described in the goals of the GiMoDig project
(Sarjakoski et al., 2002a,b), the Global Schema should
Inte
For
the
Swe
attri
The
not
seen
Thei
Cros:
Edg:
diffe
data:
of tl
diffe
a mu
4.3
Afte
that
have
these
follo
selec
deal
datas
"Tour
whic
Nisse
requi
small
Sche
types
featu
can E
denoi
an ec
lacks
GiMc
comn
Suppe
count