The International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences. Voi. XXXVII. Part B7. Beijing 2008 
144 
Figure 3. Point-based classification result 
Figure 4. Classification result using segmentation result 
4. CONCLUSION 
Decomposition and Algorithm Evaluation. Proc.of SPIE. 6790, 
pp. 101 - 108. 
[6] . Cloude S., 1996. A review of target decomposition 
theorems in radar polarimetry. Geoscience and Remote 
Sensing, 34(5), pp. 498-518. 
[7] . Ferro-Famil., 2001. Unsupervised Classification of 
Multifrequency and Fully Polarimetrie SAR Images 
Based on the H/A/Alpha-Wishart Classifier. Geoscience 
and Remote Sensing. 39(11), pp. 2332-2342. 
ACKNOWLEDGEMENTS 
Thanks for Supporting BY the 973 program of People's 
Republic of China under Grant 2006CB701303. 
In order to overcome the disadvantages of RGW method, and 
make it applicable to PolSAR data classification, one novel 
edge-detection algorithm based segmentation is proposed in this 
paper. In this paper a new object-oriented classification 
algorithm is proposed and the features and improvements of the 
new method are concluded. Also the algorithm to segment is 
applied AIRSAR PolSAR data. In order to verify the validity of 
the segmentation result, a simple classification based on the 
segment results was performed and the result was compared 
with that got using point based classification method, and the 
superior classification results got by new method demonstrated 
the effectiveness of the new segment method. 
REFERENCES 
[1] . White, R. G., 1991. Change Detection in SAR Imagery. Int. 
J. Remote Sensing, 12, pp. 339-360 
[2] . White, R. G., 1986. Low Level Segmentation of Noisy 
Imagery. DRA Memorandum, 3900. 
[3] . Oliver. C., 1998. Understanding synthetic aperture radar 
images. Artech House, pp. 197-201. 
[4] . Caves. R. G., 1996. Multi-Channel SAR Segmentation: 
Algorithms and Applications. The Institution of Electrical 
Engineers. 
[5] . Guo. Hd., 2000. Theories and Applications of Radar 
for Earth Observation. Beijing Science Press, pp. 101- 
105. 
Yang. J. 2007. Unsupervised Classification of Polarimetrie SAR 
Images Using Complex Wishart Distribution Based on H/a