International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences, Vol XXXV ‚ Part B2. Istanhul 2004
Automatic Green Red
Human Operator
Green 65.7 94 32.0 %
Red 0.76 % 1.53 %
Table 3. Statistics for road verification. The results are based on
the number of ATKIS road objects within rural areas. The
test area comprises 14 images each of a size of 2 km x 2 km.
Figure 4 depicts an example showing a development area that
has been detected automatically. The statistics for the
verification of built-up areas are revealed by Table 4. It
indicates a comparatively high number of false positives (cf.
Table 1). Most of these cases originate from single buildings
only covering a small part, e.g. less than 10 %, of the parcel.
We want to solve this by an adaptation of the procedure to these
situations.
Figure 4. Example for the verification of built-up areas.
Orthoimage and ATKIS DLMBasis (left), automatic
detection of errors (right).
Automatic Green Red
Human Operator
Green 69.5 % 21.2%
Red 6.2 % 3.1%
Table 4. Statistics for the verification of built-up arcas based on
14 images cach of a size of 2 km x 2 km. The percentage
relates to the number of ATKIS objects tested.
6. CONCLUSIONS AND OUTLOOK
The automated verification method as described in this paper is
characterised by an efficient workflow of automatic procedures
and a final interactive inspection of situations which turned out
to be critical in an automatic self-diagnostics. We expect to
achieve further improvement of efficiency and of the number of
undetected errors since some problems have been identified and
solutions are in process. Additionally we plan to test our
method with images from high-resolution satellites.
REFERENCES
Baltsavias, E., Grün, A., van Gool, L. (eds.), 2001. Automatic
Extraction of Man-Made Objects from Aerial and Space Images
740
(III. A.A. Balkema Publishers. Lisse / Abingdon / Exton (PA) /
Tokyo.
Bückner, J., Pahl, M., Stahlhut, O., Liedtke. C-E. 2002. A
knowledge-based system for context dependent evaluation of
remote sensing data. In: L. J. v. Gool (ed.), DAGM-Symposium,
Lecture Notes in Computer Science, Vol. 2449, Springer,
Zurich, Switzerland, pp. 58—65.
Busch, A., Willrich, F., 2002. System design for automated
quality control of geodata by integration of GIS an imagery.. In:
International Archives of Photogrammetry and Remote Sensing,
Vol. XXXIV, Part 2, Commission II, pp. 53—58.
Gerke, M., 2004. Quality assessment of road databases using
aerial imagery. In: /nternational Archives of Photogrammetry
and Remote Sensing, Vol. XXXV. to appear at XX. ISPRS
congress in Istanbul.
Gerke, M., Butenuth, M., Heipke, C., Willrich, F., 2004. Graph
supported verification of road databases. /SPRS Journal of
Photogrammetry and Remote Sensing, 58(3-4), pp. 152-165.
Gimel’farb, G. L., 1997. Gibbs fields with multiple pairwise
pixel interactions for texture simulation and segmentation.
Rapport de recherche RR-3202, INRIA, Sophia Antipolis
France.
Heipke, C, ^Pakzad, K; Willrich, F., Peled, A.. 2004.
Integration of geodata and imagery for automated update of
spatial databases. ISPRS Journal of Photogrammetry and
Remote Sensing, 58(3-4), pp. 127—128.
Joos, G., 2000. Zur Qualität von objektstrukturierten Geodaten.
Schriftenreihe des Studienganges Geodäsie und Geoinformation
der Universität der Bundeswehr München, Nr. 66.
Müller, S., Weis, M., Liedtke, C.-E. and Pahl, M., 2003.
Automatic quality surveillance of GIS data with GeoAIDA. In:
Proc. PLA: 2003 ISPRS Conference on Photogrammetric
Image Analysis, Munich, Germany, pp. 187-193.
Rehrmann, V., Priese, L., 1998. Fast and robust segmentation of
natural .color. scenes. 1a: R1. Chin, T.-C. Pong (eds.),
ACCV'98, Third Asian Conference on Computer Vision, Hong
Kong, China, Proceedings, Volume I. Lecture Notes in
Computer Science 1351 Springer, pp. 598-606.
Steger, C., 1998. An unbiased detector of curvilinear structures.
IEEE Transactions on Pattern Analysis and Machine
Intelligence, 20(2). pp. 311—326.
Vosselman, G., 1996. Uncertainty in GIS supported road
extraction. In: /nternational Archives of Photogrammetry and
Remote Sensing, Vol. XXXI, B3, pp. 909-917.
Walter, V., 2004. Object-based classification of remote sensing
data for change detection. /SPRS Journal of Photogrammetry
and Remote Sensing, 58(3-4), pp. 225-165.
Wiedemann, C., 2002. Extraktion von StraBennetzen aus
optischen Satellitenbildern. Deutsche Geoditische Kommission.
Reihe C, Dissertationen, Heft Nr. 551.
Wiedemann, C.. Ebner, H., 2000. Automatic completion and
evaluation of road networks. In: International Archives of
Photogrammetry and Remote Sensing, Vol. XXXIII, Part B3/2,
Commission II, pp. 979-986.
KEY W(
ABSTR/
In the re
efficient |
satellite s
acquisitic
Swiss Fe
networks
on rural :
have a m
exists, it
extensive
orthoima,
on the pe
qualitativ
correctne
of 0.4 m
In moder
stored in
containin
database
Mapping
generate
reality an
various e
accurate
roads ang
and fleet
systems,
eic. In 2
manufact
evaluate
(Intellige
as well ¢
and
(Rp:
Also in
administr
market s
Data Sol
the Et
Infrastruc
elevation
agreed to
