基于改进型互信息的遥感图像配准方法  

A Remote Sensing Image Registration Algorithm Basing on the Improved Mutual Information

作　　者：韦春荣[1] 何楚[2] 

机构地区：[1]广西师范大学电子工程学院,广西桂林541004 [2]武汉大学电子信息学院,湖北武汉430072

出　　处：《广西师范大学学报（自然科学版）》2014年第2期20-25,共6页Journal of Guangxi Normal University:Natural Science Edition

基　　金：国家自然科学基金资助项目(60702041);广西教育厅科研基金资助项目(200911MS51)

摘　　要：互信息作为图像配准的相似性测度函数,同等程度地包含了待配准图像重叠区域中感兴趣信息和冗余信息。本文将图像中感兴趣区域的互信息引入到图像相似性测度函数中,给出一种互信息和感兴趣区域互信息相结合的新的图像配准相似性测度函数,再基于该测度函数实现对遥感图像的配准。实验结果表明本文算法在提高配准精度上的有效性。As a similarity measure function for image registration, mutual information equably containsthe information of the interesting region and the other region of the overlap area between the registrationimages pair. In this paper, a new similarity measure function which combines both the mutual informa-tion of the interesting region and the tradition mutual information of the images is provided to direct theimage registration process. The experimental results of remote image registration show that the registra-tion algorithm based on the new similarity measure function is effectively in improving the registrationaccuracy.

关 键 词：互信息 感兴趣区域 遥感图像配准 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]