The International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences. Vol. XXXVII. Part B3b. Beijing 2008 
608 
6. CONCLUSION 
Compared to the traditional methods, our method has an 
advantage in detecting roads from SAR images with higher 
accuracy. The homogeneous factor of road and its background 
is introduced into the algorithm, as a result, the extraction of 
road edge becomes smoother. On the other hand, the extraction 
doesn't become computational complexity along with this 
introduction. Hence, the algorithm proposed in this paper is suit 
for the road extraction from SAR images. 
REFERENCES 
Chen, Y., 2003. Digital Photogrammetry for Remote Sensing 
Image. Tongji University, Shanghai, pp. 150-155. 
Lee, J., 1989. Segmentation of SAR images. IEEE Transaction 
on GeoScience and Remote Sensing, 27(6), pp. 674-680. 
Touzi, R., 1988. A statistical and geometrical edge detector for 
SAR images. IEEE Transaction on GeoScience and Remote 
Sensing, 26(6), pp. 764-773. 
Xie, F., 2007. Road extraction from RS imagery based on 
wavelet and mathematical morphology. Computer Engineering 
and Applications, 43(22), pp. 241-243. 
Zhang, S., 2007. Edge detection for SAR image based on Least 
Square Support Vector Machine. Measurement & Control 
technology, 26(11), pp. 61-63. 
Geling, G., 1993. An edge detection operator for SAR images. 
Canadian Conference on Electrical and Computer Engineering , 
2, pp. 707-709.