改进IR-MAD算法的高分光学卫星影像相对辐射归一化方法  

Relative Radiometric Normalization Method of High-resolution Optical Satellite Image based on An Improved IR-MAD Algorithm

作　　者：樊登辉 李利伟 申茜 张浩 程钢[1] FAN Denghui;LI Liwei;SHEN Qian;ZHANG Hao;CHENG Gang(School of Surveying and Land Information Engineering,Henan Polytechnic University,Jiaozuo 454000,China;Aerospace Information Research Institute,Chinese Academy of Sciences,Beijing 100094,China)

机构地区：[1]河南理工大学测绘与国土信息工程学院,河南焦作454000 [2]中国科学院空天信息创新研究院,北京100094

出　　处：《遥感技术与应用》2024年第4期859-866,共8页Remote Sensing Technology and Application

基　　金：国家重点研发计划项目(2021YFB3901101);国家自然科学基金项目(41971327)。

摘　　要：相对辐射归一化是高分光学卫星影像地表反射率产品生产的主要手段之一,基于迭代加权多元变化检测(Iteratively Reweighted Multivariate Alteration Detection,简称IR-MAD)的高分光学卫星影像相对辐射归一化方法是当前应用最为广泛的方法之一。但是,经典IR-MAD方法仅适用于参考影像与待校正影像之间地物变化不大的情况,当地物变化较大时该方法失效。通过引入伪不变区域约束机制,提出了一种优化的IR-MAD算法,主要思路是借助二元高斯混合聚类方法提取影像之间不变区域,进而提升伪不变点提取的质量,使地物变化较大时仍然可以输出有效的辐射归一化结果。以高分二号影像为待校正影像,哨兵二号影像为参考影像,采用3组代表性影像进行实验验证,并与经典IR-MAD方法相比,结果表明:①该方法提取的伪不变点主要集中在不变区域,空间分布更加合理;②该方法在代表一般变化的数据集1中,平均决定系数提高8.8%,在代表强烈变化的数据集2、3中,该方法平均决定系数分别为95%、89%,而经典方法低于50%甚至是负值;③该方法处理结果影像中典型地物的反射光谱信息与参考影像更加接近。因此,该方法为高分光学卫星影像相对辐射归一化提供了新的有效途径。Relative radiometric normalization is one of the main means of producing surface reflectance products from high-definition optical satellite images,the relative radiometric normalization method based on Iteratively Reweighted multivariate alteration detection is currently the most widely used method for high-resolution opti⁃cal satellite image.However,the classical IR-MAD method is only applicable when there is little change in fea⁃tures between the reference image and the image to be corrected,and the method fails when there is a large change in features.In this paper,an improved IR-MAD algorithm is proposed by introducing a pseudo-invari⁃ant region constraint mechanism.The main idea is to use a Gaussian mixture model to extract the invariant re⁃gions between images,thus improving the quality of the pseudo-invariant points,so that effective radiometric normalization results can still be output when the feature changes a lot.The experiment was conducted using three representative sets of images,with GF-2 as the image to be corrected and Sentinel-2 as the reference im⁃age.The results show that compared to the classical IR-MAD method,(1)The pseudo-invariant points extract⁃ed by our method are mainly concentrated in the invariant regions,and the spatial distribution is more reason⁃able;(2)In dataset 1 representing general changes,the average coefficient of determination of our method is in⁃creased by 8.8%,while in dataset 2 and 3 representing strong changes,the average coefficient of determination of our method is 95%and 89%,respectively,while the traditional method is lower than 50%or even negative;(3)The spectral reflectance information of typical features in the processed image by our method is closer to the reference image.Therefore,our method provides a new and effective approach for relative radiometric normal⁃ization of high-resolution optical satellite images in complex situations.

关 键 词：相对辐射归一化 迭代加权多元变化检测 高分辨率光学卫星影像 伪不变区域 

分 类 号：P237[天文地球—摄影测量与遥感]