International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences, Vol XXXV, Part B4. Istanbul 2004
temporarily discarded matchings, which are in conflict with it,
have to be definitely discarded.
For line objects this iterative approach leads to good results for
the criterion “similarity of length”, which means that a treshold
is defined for the maximum of the two quotients of the length.
3.3.3 Manual Selection
The automatic selection procedure leaves a set of possible
matchings which cannot automatically be confirmed or
discarded. These are, for example, pairs of matchings, which
hold the given set of quality tresholds and are in conflict with
each other. In such cases the decision for confirming or
discarding must be left to a human operator.
The operator must be provided with an interface, which helps
him making a decision. Therefore a graphical user interface is
needed, which shows the uncertain matchings in their context
and lets the user confirm or discard.
We have implemented an extension for the visualizer
GlSVisual, which was formerly developed at our institute,
which provides the user with these features and an interface for
administration of the federated database. The interface provides
firstly a graphical user interface for capturing and changing the
needed meta data and parameters, for example for the semantic
classification. Then there is the possibility to register and
parameterize procedures for finding possible matchings (as an
alternative to the buffer growing) as well as for the selection
procedures. These are procedures implemented in PL/SQL.
For the manual selection the functions for marking pairs of
objects, in this context the matching pairs, and calling database
procedures on this pair, were implemented. The operator
therefore can choose one or more pairs and afterwards discard
or confirm them with respect to the conflict rules. If configured
by the operator, the process of confirmation of non-conflicting
matchings, will start after each manual confirmation.
4. FUTURE WORK
Now, that we have a framework for generating a multi-
resolution database and some methods to match line objects, we
are focussing on tuning the matching process and augmenting
the degree of automation in the selection process, which means
to experiment with different parameterizations for the existing
procedures as well as developing new procedures.
Another focus has to be set on the development of region
matching algorithm respectively the integration of exiisting
ones.
5. REFERENCES
Conrad, S., 1997.
Verlag, Berlin.
Föderierte Datenbanksysteme. Springer
Devogele, T., Parent, C., Spaccapietra, S., 1998. On Spatial
Database Integration. International Journal of Geographical
Information Science, 12(4), pp. 335-352.
Kleiner, C., Lipeck, U.W., 2001. Enabling Geographic Data
with Object-Relational Databases. In: A. Heuer et al.,
176
Datenbanksysteme in Büro, Technik und Wissenschaft — 9. GI-
Fachtagung BTW 2001, Springer Verlag, Berlin, pp. 127-143.
Lipeck, U.W., Mantel, D., 2004. Datenbankgestütztes Matching
von Kartenobjekten. To appear in: Mitteilungen des
Bundesamtes für Kartographie und Geodäsie, Bundesamt für
Kartographie und Geodäsie, Frankfurt am Main.
Mantel, D., 2002. Konzeption eines Föderierungsdienstes für
geographische Datenbanken. Master Thesis, University of
Hannover, Germany.
Sester, M., Anders, K.-H., Walter, V., 1999. Linking Objects of
Different Spatial Datasets by Integration and Aggregation.
Geolnformatica, 2(4), pp. 335-358.
Sester, M., 2000.
digitalen räumlichen Datenbeständen.
University of Stuttgart, Germany.
Maßstabsabhängige Darstellungen in
Habilitation Thesis,
Tiedge M., Lipeck, U. and Mantel, D., 2004. Design of a
Database System for Linking Geoscientific Data.
Geotechnologien Science Report "Information Systems in Earth
Management", No. 4, Koordinierungsbüro Geotechnologien,
Potsdam, 2004, pp. 83-87.
Walter, V., 1997. Zuordnung von raumbezogenen Daten — am
Beispiel der Datenmodelle ATKIS und GDF. Ph.D. Thesis,
University of Stuttgart, Germany.
6. ACKNOWLEDGEMENTS
This work is funded by the German Federal Agency for
Cartography and Geodesy as part of the WiPKA (Wissens-
basierter Photogrammetrisch-Kartographischer Arbeitsplatz)
project and by the Federal Ministry for Education and Research
(BMBF) and the German Council (DFG) as part of the
Geotechnologien project under contract no. 30F0374A.
KEY
ABS
The
on tl
Exte
fram
servi
out €
quer
enco
prese
Integ
estab
over
techr
platf
tradi!
settir
Sim
Asn
focu:
relate
spati:
data
areas
2004
a wi
deali
The 1
XML
The
scher
Thes
issue
mobi
In thi
motis
intro
techn
a prc
four-i
Servi
(GiM
