一种基于结构相似度的IHS与小波变换相结合的遥感影像融合算法  

An IHS and wavelet integrated fusion method based on structural similarity

作　　者：王晓艳[1] 刘勇[1] 蒋志勇[1] 

机构地区：[1]兰州大学资源环境学院,兰州730000

出　　处：《兰州大学学报（自然科学版）》2011年第5期1-6,共6页Journal of Lanzhou University(Natural Sciences)

基　　金：中央高校基本科研业务费专项资金自由探索项目(lzujbky-2010-178;lzujbky-2009-71)

摘　　要：针对分别使用IHS变换和小波技术进行融合实验时有较大的色彩畸变的问题,提出了一种基于结构相似度的IHS与小波变换相结合的融合算法.首先对多光谱影像进行IHS变换,对多光谱影像的亮度分量Ⅰ和全色波段影像进行小波分解;然后采用二者高频系数绝对值较大者作为新的Ⅰ分量的高频系数,而新的Ⅰ分量的低频系数由二者的低频系数进行加权平均获得,权值通过计算Ⅰ与全色波段的SSIM来自适应地确定.实验结果表明该算法在提高影像分辨率的同时,能很好地保持影像的光谱特征.The fused image always display significant color deviation from the original image when IHS transform and wavelet transform are used separately. In this paper, a new IHS and wavelet transform integrate fusion method based on structural similarity （SSIM） was proposed. First, we transformed the multispectral image from RGB to IHS color space, and then decomposed both the panchromatic image and the intensity component into wavelet planes. In high frequency sub-band, the coefficients whose absolute value was larger were chosen as the high frequency coefficients of new intensity component. In low frequency sub-band, the low frequency coefficients of the new intensity component were the weighted average of the panchromatic and the intensity component, and the weight value was determined adaptively by the structural similarity（SSIM） between them. The experiment results show that this algorithm can preserve the spectral characteristics of the multispectral image, thus improving the resolution of the image.

关 键 词：遥感 图像融合 IHS变换 SSIM 小波变换 

分 类 号：P91[天文地球—自然地理学] TP753[自动化与计算机技术—检测技术与自动化装置]