429 
a) Image with gauss noise 
b) MRED algorithm 
c) Sobel algorithm 
d) Pal.King algorithm e) FGED algorithm f) WFCE algorithm 
Figure4 Contrastive test of five algorithms in 10% gauss noise 
i appears over 
iject is cannot 
rED algorithm is 
affected by noise 
gauss noise, the 
tuns in different 
Compare the detect result, we find that the MRFD algorithm 
have an obviously relatively superiority of antinoise capability 
in edge detection of multispectral image. 
An edge detection algorithm for multispectral RS image 
(MRED) is proposed based on the detailed analysis of the 
characters of multispectral image. Through the multi 
dimensional cloud-space mapping model the objects in image 
can be mapped to the cloud-space, the fuzzy feature matrix that 
covered by multi-dimensional edge cloud is extracted by 
Boolean calculation between intersectant clouds. Calculating 
image fuzzy division entropy by fuzzy feature matrix of each 
sub-cloud space, bring stochastic influence of image into 
solution of entropy and use fuzzy division entropy repeatedly to 
find the best result in membership. The edge map with 
preferable precision can be obtain by integrate the detect result 
of every sub-cloud space. MRED algorithm considering fuzzy 
features of remote sensing image and discussing the connatural 
stochastic of image at the same time, the contrastive test is 
proved that this algorithm is more effective than contrastive 
algorithm on detection quality and antinoise. 
ACKNOWLEDGMENTS 
The work is supported by Chongqing Nature Science Fund 
(CSTC 2007BB2392). 
REFERENCES 
1. Baraldi, F. Parmiggiani. A Neural Network for 
Unsupervised Categorization of Multi-valued Input Patterns: An 
Application to Satellite Image Clustering, IEEE Transactions 
on Geo-Science and Remote Sensing, 1995, 33(2): 305-316 
2. Mannan, J. Roy and A. K. Ray, 1998, Fuzzy Artmap 
Supervised classification of multi-spectral remotely-sensed 
images, Int. J. Remote Sensing, 19(4): 767-774. 
3. R. C. Gonzalez and R. E. Woods, Digital Image Processing, 
Addison-Wesley, Reading, MA, 1992 (Sec. 4.6.2) 
4. ZHOU Chenghu, LUO Jiancheng, Geographic 
Comprehend and Analysis of Remote Sensing Image, science 
Press, Bei Jing, 1999. 
5. ZHANG Rong, LIU Zhengkai, ZHAN Kai. Compression 
of multispectral images by wavelets transition. Transactions on 
Geo-Science and Remote Sensing, 2000, 4(2): 100-105. 
6. P. L. Dragotti, G. Poggi, A. R. P. Ragozini. Compression 
of multispectral images by three-dimensional SPIHT algorithm. 
IEEE Trans. Geosci. Remote Sensing, 2000, 38(1): 416-428. 
7. A. A. Kassim, W. S. Lee. Embedded color image coding 
using SPIHT with partially linked spatiall orientateon tress. 
IEEE Trans. Circuits Syst. Video Technol., 2003, 13(2): 203- 
206 
8. YAN Jingwen, SHEN Guiming. Compression of three- 
dimensional multispectral images by KLT/WT/WTVQ. Journal 
of XiaMeng University, 2001,40(5): 1051-1055. 
9. K. Karhunen, Uber Linear Methoden in der 
Wahrscheinlich-Keitsrechnung, Ann. Acad. Sci. Fennicae, Ser. 
A. I. 37, 1974 
10. M. Lève, Fonctions Aléatoires de Second Ordre, In P. 
Lévy, Proessus Stochastiques et Mouvement Brownen, 
Hermann, Paris, 1948 
11. Xue L.X. Wang Z.C. Li Y.SH. Wang L.L. “Fuzzy Edge 
Detection Based on Cloud Model”, Journal of Southwest 
Jiaotong University,2006,41 ( 1 ):85-90. 
12. WANG Shuliang, SHI Wenzhong, LI Deyi, LI Deren, 
WANG Xinzhou, A method of spatial data mining dealing with 
randomness and fuzziness. Proceedings of the Second 
International Symposium on spatial Data Quality, edited by