一种基于结构相似度的IHS变换融合算法  

An IHS Fusion Method based on Structural Similarity

作　　者：王晓艳[1] 刘勇[1] 蒋志勇[1] 

机构地区：[1]兰州大学资源与环境学院,甘肃兰州730000

出　　处：《遥感技术与应用》2011年第5期670-676,共7页Remote Sensing Technology and Application

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)

摘　　要：由于IHS色彩空间表现的颜色更加符合人眼的视觉规律,因此,IHS变换在遥感影像融合中被广泛应用。针对传统的IHS变换融合算法进行融合实验时有较大的色彩畸变问题,提出了一种基于结构相似度(Strucral Similarity SSIM)的IHS变换融合算法。对影像进行IHS变换之后,计算原始多光谱影像I分量与全色波段影像的SSIM矩阵,并由该SSIM矩阵确定对应不同区域的新的亮度分量I。实验结果表明,该算法在增强影像空间分辨率的同时,能很好地保持其光谱特征。IHS（Intensity-Hue-Saturation） color space is more consistent with human visual system than RGB color space,so IHS transform has been used widely in image fusion.The most significant problem of the traditional IHS transform fusion method is that the fused image usually has a notable deviation in visual appearance and in spectral values from the original image.In this paper,we propose a new IHS transform fusion method based on Structural Similarity（SSIM）.First,the multispectral image was transform from RGB color space to IHS color space,then we calculate the SSIM matrix between the intensity component of the multispectral image and the panchromatic image.The new intensity component is determined based on the SSIM.Experimental results indicate that this method is effective in preserving spectral and spatial information.

关 键 词：图像融合 IHS变换 SSIM 遥感 

分 类 号：TP753[自动化与计算机技术—检测技术与自动化装置]