基于目标匹配的遥感图像变化检测方法  

Change detection algorithm for remote sensing images based on object matching

作　　者：苏娟[1] 林行刚[1] 刘代志[2] 

机构地区：[1]清华大学电子工程系,北京100084 [2]第二炮兵工程学院602室,西安710025

出　　处：《清华大学学报（自然科学版）》2007年第10期1610-1613,共4页Journal of Tsinghua University(Science and Technology)

基　　金：国家自然科学基金资助项目(60472028);国家教育部博士点基金项目(20040003015)

摘　　要：针对传统的像素级遥感图像变化检测方法中检测精度严重依赖于图像配准、辐射校正和差异图像阈值选取的问题,提出了一种基于目标匹配的目标级遥感图像变化检测方法。该方法直接作用于2幅未经配准和辐射校正的不同时相遥感图像,利用目标的区域不变矩与目标之间的角度关系对目标进行匹配,将不能匹配的目标作为变化目标,利用匹配目标对2幅图进行配准,以同时实现图像的配准与变化检测,并用配准结果对变化检测结果进行修正。实验结果表明,该方法具有较好的定性检测性能。The detection accuracy of traditional pixel-level change detection algorithms is seriously affected by the registration accuracy, the radiometric correction accuracy, and the classification threshold for difference image. An change detection algorithm based on object matching was developed to compare two images without any registrations and radiometric corrections. Changes of man made objects are detected based on the similarities of moment invariants and angular relationships between objects. The detected unchanged objects are used to register two images, with the change detection results verified by registration results. Test results indicate that the algorithm gives good qualitative detection.

关 键 词：遥感图像 变化检测 人造目标 目标匹配 图像配准 

分 类 号：TP75[自动化与计算机技术—检测技术与自动化装置]