关于SAR图像配准技术综述.pdf

第21卷第1 1期 2013年1 1月光学精密工程()ptics and Precision Engineering 3 文章编号 1004 924X(20I 3)1 I一2960—1 3 遥感图像配准技术综述余先川”,吕中华,胡丹(北京师范大学信息科学与技术学院,北京100875) 摘要:遥感图像配准是图像融合、多光谱分类、环境监测和图像镶嵌等不可缺少的步骤。、基于图像特征的配准、基于混合模型的配准和基于物理模型的配准四类;,重点概述了基于特征配准中的局部不变特征变换算法。评述了国内外遥感图像配准的发展现状;指出了遥感图像配准技术中存在的问题,即多源遥感图像的配准、遥感图像配准的实时性、遥感图像的非线性配准和遥感图像配准的精度评价,最后展望了遥感图像配准技术的发展前景。关键词:图像配准;遥感图像;特征提取;物理模型中图分类号:7FP751 文献标识码:A doi:/() 3211 Review ofremote sensing image r stration techniques YU Xian—chuan+,I,U Zhong—hua。HU Dan (College oJ’hzformation S(‘iem’P&五(,hnology,Beijing NOFllldl University。*C(JrrF—po”dj 729authol4,E-mail:’hl,lgt?******@163. ]3eijing 100875。Chb?“) Abstract:Remote sensing image registration is an indispensable part for remote sensing image fusion, multispectral classification,environmental monitoring,image mosaicing and SO thispaper,the important and latest registration methods forremote sensing are discussed and are divided into four types,including area—based methods,feature—based methods,hybrid—based model methods and physi— cally—based model ,the classicalgorithms ofeach type are analyzed respectively,and theiradvantages and ings are also scale invariant feature transform algorithms are mainly ,the difficultiesofremote sensing image registration techniques are summarized,including the multi—source remote sensing image registration,the real—time registration ofremote sensing image,the nonlinear registration ofremote sensing image and the accuracy evalua— tion ofremote sensing image ,the prospects ofimage registration are pointed out. Key words:image registration;remote sensing image;feat ure extraction;physical model 引—L 图像配准是指在不同时段,对同一场景从不同视角使用相同或不同的传感器拍摄的有重叠区域图像进行几何校准的过程‘。2『。图像配准技术是近年来发展迅速的图像处理技术之一,主要用于军事、遥感、医学、计算机视觉等领域。国外从收稿日期:201 3—0225;修订日期:201 3—04—05. 基金项目:国家自然科学基金资助项目( 272359);北京市自然科学基金资助项目(