A relative radiometric correction method based on geographically weighted regression model

Chengfeng Luo, Weilong Wu,Haoyan You, Jiao Wang

PROCEEDINGS OF THE 2013 THE INTERNATIONAL CONFERENCE ON REMOTE SENSING, ENVIRONMENT AND TRANSPORTATION ENGINEERING (RSETE 2013)(2013)

引用 23|浏览1
暂无评分
摘要
to detect the change in the terrain using multitemporal images has becoming one of the important applications of remote sensing technology. In order to receive a result with high accuracy, the relative radiometric correction among images must be done before the detection. The surface radiation is affected by spatial correlation among the surface objects. This study introduced the geographically weighted regression to the radiometric correction process, and proposed a radiometric correction method based on the geographically weighted regression model. The method includes three main steps. Firstly, iterative weighted multivariate change detection is used to select the invariable pixels as samples. Secondly, radiation correction linear model is built at each sampling point based on geographically weighted regression. Finally the radiometric correction values of target points are calculated with the model of closest point. In the test, using the proposed method a good visual effect can be received. And the precision evaluation indexes are better than those of results from the orthogonal regression radiometric correction. Especially, the information entropy index is almost as twice much as the original image and that of orthogonal regression radiometric correction. It can be concluded that the proposed method in this paper can ensure the radiometric correction visual effect and enhance the details of performance ability of images at the same time.
更多
查看译文
关键词
relative radiometric correction,geographically weighted regression (GWR),information entropy,invariant features points
AI 理解论文
溯源树
样例
生成溯源树,研究论文发展脉络
Chat Paper
正在生成论文摘要