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.
杨存建, 许珺, 张增祥. SAR和TM图像主成分变换融合中不同主分量替换的比较[J]. 国土资源遥感, 2001, 13(3): 30-35.
YANG Cun-jian, XU Jun, ZHANG Zeng-xiang . COMPARISON OF SUBSTITUTING THE DIFFERENT PRINCIPLE COMPONENT IN THE FUSION OF TM AND SAR IMAGE PRINCIPLE COMPONENT CHANGE. REMOTE SENSING FOR LAND & RESOURCES, 2001, 13(3): 30-35.
[1] 游先祥,杨晓明.应用遥感信息复合方法的森林分类和动态监测研究［J］.环境遥感,1995,10(2):97-106.[2] 唐伶俐,江平,潘京明,等.SAR与TM图像复合处理技术的初步研究［J］.遥感信息,1995,9(4):15-16.[3] 方红亮,杨晓梅,杜云艳.运用主成分变换-逆变换对ADOES AVNIR XS和PAN进行复合的研究［J］.遥感技术与应用,1998,13(3):48-53.[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.[8] 刘同明.基于证据理论模糊推理的多传感器信息融合海上目标识别［J］.模糊识别与人工智能,1999,12(1):25-31.[9] 贾永红,李德人,等.四种HIS变换用于SAR与TM影像复合的比较［J］.遥感学报,1998,2(2):103-106.[10] 贾永红.TM和SAR影像主分量变换融合法［J］.遥感技术与应用,1998,13(1):46-49.
2001, Vol. 13 
