Abstract Principal component analysis(PCA) is one of the commonly_used methods to fuse Multisensor image data. Its procedure is to replace the first component with the high_resolution image, then to get the fused image by reversing PCA transform. Nevertheless, the first component contains more information than other components, and there will be much information lost if it is replaced. In this paper,TM2,TM3,TM4,TM5 and TM7 were analyzed by principle component transform, Radarsat SAR was used to replace the first, second, third, forth and fifth component of PCA respectively. Five different results were acquired by reversing PCA transform. The standard deviation and information entropy of five images were applied to analyze the fusion effect. The result shows that the fused images by replacing the fourth and fifth component contain more information than that by replacing the first component and can raise the separability between different types or classifications. But the difference between the fused images by replacing the fourth and the fifth component is not remarkable.
HUANG Shu-Chun,ZHANG Yuan-Ping,JIANG Feng. COMPARISON OF SUBSTITUTING THE DIFFERENT PRINCIPLE COMPONENT IN THE FUSION OF TM AND SAR IMAGE PRINCIPLE COMPONENT CHANGE[J]. REMOTE SENSING FOR LAND & RESOURCES,
2001, 13(3): 30-35.
[4] Wald I, Ranchin T, Mangolini M. Fusion of Satellite Images of Different Spatial Resolutions:Assessing the Quality of Resulting Images［J］. PE&RS, 1997,63(6):691-699.
[5] 孙家柄,刘继琳,李军.多源遥感影像融合［J］.遥感学报,1998,2(1):47-50.
[6] Pohel C, Van Genderen J L. Multisensor Image Fusion in Remote Sensing: concepts, methods and applications［J］. INT. J. Remote Sensing, 1998, 19(5): 823-854.
[7] Zhou J, Civcot D L, Silander J A. A Wavelet Transform Method to Merge Landsat TM and SPOT panchromatic Data［J］. INT. J. Remote Sensing, 1998, 19(4): 743-757.
2001,
Vol. 13
