基于遗传算法与仿射变换的遥感图像配准  

Remote-sensing Image Registration via Genetic Algorithm and Affine Transform

作　　者：张斌[1] 周杨[1] 陈书文 王水花[3] 

机构地区：[1]江苏省辐射环境监测管理站 [2]江苏省辐射环境保护咨询中心 [3]东南大学信息科学与工程学院,南京210019

出　　处：《科学技术与工程》2010年第21期5168-5170,共3页Science Technology and Engineering

基　　金：国家自然科学基金(60473065;60572063;60872075);高等学校科技创新工程重大项目培育资金项目(706028);江苏省自然科学基金(BK2007103)资助

摘　　要：为了更好地实现遥感图像配准,提出一种新的模型,将传统的平移旋转变换推广到仿射变换,采用互信息测度衡量两幅图像之间的差异,并采用遗传算法寻找模型的最优值。实验表明算法有效,能够在较短的迭代次数内收敛到问题的最优解。In order to realize image registration more efficiently,a novel model was proposed.Firstly,traditional translation and rotation transform was generalized to affine transform.Secondly,the mutual information was used to measure the difference between two images.Finally,genetic algorithm was used to find the global optimal point of the model.Experiments demonstrate that this proposed method is effective,moreover,it can converge to the global optimal solution within small epochs.

关 键 词：图像配准 遗传算法 仿射变换 

分 类 号：TN911.73[电子电信—通信与信息系统]