基于IHS变换和主成分变换的遥感影像融合  

Remote sensing image fusion based on IHS transform and principal component analysis transform

作　　者：武文波[1] 康停军[2] 姚静[1] 

机构地区：[1]辽宁工程技术大学测绘与地理科学学院,辽宁阜新123000 [2]中山大学遥感与地理信息工程系,广东广州510275

出　　处：《辽宁工程技术大学学报（自然科学版）》2009年第1期28-31,共4页Journal of Liaoning Technical University (Natural Science)

基　　金：教育部博士点基金资助项目(20050147002);辽宁省自然科学基金资助项目(20042175)

摘　　要：针对IHS变换融合影像时存在较严重的光谱失真现象,利用主成分变换对IHS变换法进行了改进。新方法首先对多光谱影像做IHS变换得到亮度I,色度H,饱和度S三个分量,然后用I分量和高分辨率全色影像做主成分变换,并提取第一主分量,并以I为标准进行直方图匹配;将匹配后的影像与H,S进行IHS反变换得到新的多光谱图像。主观视觉分析和客观参数表明,该方法不仅很好的保留了影像的光谱信息,而且兼顾了地物细节能力的表达。In view of the spectrum distortion of IHS transformin, an improved remote sensing image fusion method is presented based on IHS transforming and Principal Component Analysis （PCA） Transforming. Firstly, the multi-spectral image is transformed by IHS transform, and I,H,S components are obtained ;Secondly, the panchromatic image and I component are merged with PCA transforming. And the first component is obtained and the histogram of the first component is matched according to I. and I was replaced by the matched data; finally, the fused image is obtained by inverse IHS transform. The new method preserves spectral information of the original multispectral image well and enhances spatial information largely greatly.

关 键 词：遥感影像融合 IHS变换 主成分变换 

分 类 号：TP751[自动化与计算机技术—检测技术与自动化装置]