一种遥感影像云层遮挡消除方法  

A Method of Eliminating Clouds Covering in Remote Sensing Imagery

作　　者：简剑锋[1] 李新涛[2] 王任享[2] 周利华[1] 尹忠海[1] 

机构地区：[1]西安电子科技大学计算机网络与信息安全教育部重点实验室,西安710071 [2]西安测绘研究所,西安710054

出　　处：《光子学报》2009年第11期3044-3048,共5页Acta Photonica Sinica

基　　金：国家部委型号研制基金(A200206210)资助

摘　　要：针对光学遥感影像中常见的云层遮挡问题提出了一种去除云层的处理方法,采用直方图规定化来减小原始影像与替换影像间的灰度差异,采用特征点提取、整体松弛匹配以及小面元三角网纠正方法,解决了原始影像与替换影像之间的配准、纠正问题,同时利用羽化处理技术保证了拼接线附近影像灰度的平滑变化.试验结果表明:利用同一区域、不同时相、不同传感器平台的遥感影像来替换原始影像中的云层遮挡区域,可以有效解决遥感影像中的云层遮挡问题.To solve the problem that the remote sensing image always is covered by clouds,a method is presented to remove the clouds.The gray difference between the original image and the replacing one is minished by means of histogram standardization.The registering and the rectifying between the original image and the replacing one are resolved effectively,with the extracting feature points quickly,the global relaxation matching and the triangle network rectifying of small area element.And,the image gray smoothly changing by seam-line is solved by the technique of feathering.Experiment results show that,the area covered by cloud in the original image is replaced with the remote sensing image which is collected at the same region,at the different time and by the diferent sensor,and the issue that the remote sensing image was covered by clouds can be resolved effectively.

关 键 词：遥感影像 云层遮挡 配准 纠正 羽化 

分 类 号：P237[天文地球—摄影测量与遥感] TP751.1[天文地球—测绘科学与技术]