ISPRS, Vol.34, Part 2W2, “Dynamic and Multi-Dimensional GIS’’, Bangkok, May 23-25, 2001 
5. SUMMARY AND CONCLUSIONS 
This paper introduced a fully automatic approach to the 
registration of satellite images against vector maps. The 
proposed approach consists of two procedures: feature 
extraction for matching, and mismatch determination. Our 
approach for the extraction of matching features uses the 
indicators based on the NDVI derived from multi-spectral 
satellite images. Mismatch is determined through the use of 
voting displacements between the positions of the features and 
those of topographical objects projected onto satellite images. 
Experimental results show that our method can register satellite 
images against maps automatically and accurately. 
More experiments are needed including a wide variety of test 
areas. Several problems remain. As regards feature extraction, 
more studies on the threshold values are needed. Additionally, 
the choice of window size should also be examined. With 
regard to mismatch determination, more studies on the 
geometry conversion of rotation and expansion are needed. In 
particular, a method that handles these conversions is also 
examined. 
ACKNOWLEDGMENTS 
The authors are thankful to our research coworkers for valuable 
discussions. 
REFERENCES 
Duda, R. O. and Hart, P. E., “Use of the Hough transformation 
to detect lines and curves in pictures”, Comm. ACM, vol. 15, 
no. 1, pp. 11-15, 1972. 
Ebner, H., Eckstein, W., Heipe, C. and Mayer, H., “Automatic 
extraction of GIS objects from digital imagery", International 
Archives of Photogrammetry and Remote Sensing, vol 32 Part 
3-2W5, 1999. 
Grun, A. and Baltsavias, E. P., “Automatic extraction of man 
made objects from aerial and space images (II)", Birkhauser 
Verlag, Besel, 1997. 
Hild, H., Haala, N. and Fritsch, D., “A strategy for automatic 
image to map registration", International Archives of 
Photogrammetry and Remote Sensing, vol.33 Part B2, pp.287- 
294, 2000. 
Klang, D., “Automatic ground control point extraction for satellite 
scene orientation", ISPRS Workshop “From producer to user”, 
Boulder USA, October 1997. 
ORBIMAGE, http:/1www.orbimage.com/. 
Schilling, K. J. and Vogtle, T., “Satellite image analysis using 
integrated knowledge processing", International Archives of 
Photogrammetry and Remote Sensing, vol.31 Part B3, pp.752- 
757, 1996. 
Space Imaging, http:llwww.spaceimaging.com/. 
Vogtle, T. and Steinle, E., “3D modeling of buildings using laser 
scanning and spectral information”, International Archives of 
Photogrammetry and Remote Sensing, vol.33 Part B3, pp.927- 
934, 2000. 
Vohra, V. and Dowman, I. J., “Automatic registration of images 
to maps - the ARCHANGEL and ARMIES systems”, 
International Archives of Photogrammetry and Remote Sensing, 
vol.33 Part B2, pp.590-597, 2000.