基于曲波变换的全极化SAR数据与TM影像融合方法  

Fusion Method of SAR Polarimetric Data and TM Image Based on Curvelet Transform

作　　者：肖世忱[1,2] 廖静娟[1] 

机构地区：[1]中国科学院遥感与数字地球研究所数字地球重点实验室,北京100094 [2]中国科学院大学,北京100049

出　　处：《遥感信息》2014年第3期82-87,共6页Remote Sensing Information

基　　金：遥感科学国家重点实验室开放基金项目资助(OFSLRSS201205);中科院对地观测中心主任基金(Y2ZZ17101B)

摘　　要：针对SAR全极化数据与TM影像的信息互补特征,在IHS空间提出了一种基于二代曲波变换的影像融合算法。首先将两种数据分别从RGB空间变换到统一的IHS空间。然后对强度分量分别进行快速离散曲波变换,得到不同尺度与方向下的子带系数。对低频子带系数,根据SAR数据的梯度信息和亮度信息来确定融合权值;对不同尺度与方向下的高频子带系数,采用基于系数绝对值取大的融合规则来产生融合系数。接着对融合后的系数执行曲波逆变换得到新的强度分量。最后与TM影像的色调与饱和度分量一起执行IHS逆变换获得融合影像。实验结果表明:该算法可以有效地综合SAR数据与TM影像中的重要信息,其融合结果较采用经典的IHS变换和曲波变换所得结果在主观视觉效果和客观定量指标上均有所改善。Aiming at the information complementary characteristics between SAR polarimetric data and TM image,a novel fusion method in IHS space based on the second generation Curvelet transform was proposed.Firstly,two kinds of data were transformed from RGB space to IHS space,respectively.Then the fast discrete Curvelet transform was performed on the intensity component respectively to obtain the sub-band coefficients at different scales and in various directions.For low frequency coefficients,the fusion weights were determined by the gradient and brightness information of SAR data.While for high frequency sub-band coefficients,a fusion rule based on taking large absolute value was employed.Next,the new intensity component was obtained through the inverse Curvelet transform.Finally,the inverse IHS transform was performed to get the fusion result.Experimental results show that the proposed method can effectively integrate important information from SAR data and TM image,and produce a better result both on the subjective visual effect and objective quantitative indicators than that based on classic IHS transform method and Curvelet transform method.

关 键 词：曲波变换 SAR TM RADARSAT-2 影像融合 

分 类 号：TP751[自动化与计算机技术—检测技术与自动化装置]