一种带几何约束的大幅面遥感影像自动快速配准方法  

A Rapid Image Registration Method Based on Restricted Geometry Constraints for Large-Size Remote Sensing Image

作　　者：张谦[1] 贾永红[1,2] 吴晓良[3] 胡忠文[1] 

机构地区：[1]武汉大学遥感信息工程学院,湖北武汉430079 [2]武汉大学测绘遥感信息工程国家重点实验室,湖北武汉430079 [3]国家测绘地理信息局卫星测绘应用中心,北京100830

出　　处：《武汉大学学报（信息科学版）》2014年第1期17-21,31,共6页Geomatics and Information Science of Wuhan University

基　　金：国家科技支撑计划资助项目(2011BAB01B05);国家自然科学基金资助项目(41001260);中央高校基本科研业务费专项资金资助项目(201121302020003)~~

摘　　要：针对大幅面遥感影像的配准需求,提出了一种带有几何约束的大幅面遥感影像快速自动配准方法。该方法首先结合遥感影像空间地理信息构建影像粗配准模型;然后在粗配准模型的约束下,通过对大幅面遥感影像进行规则格网划分,结合改进的SIFT算法和并行策略实现影像精配准。实验结果表明,本文方法能提高影像的配准效率,且配准精度较高,适用于大幅面遥感影像之间的配准。Considering the registration of large-size remote sensing images, a rapid image registration method based on restricted geometry constraints is proposed. Firstly, the coarse image registration combining geography information is introduced for remote sensing image. Then, by dividing the large- size image into grid images, fine registration is done combining an improved SIFT algorithm and parallel architecture under the constraints of a coarse registration model. Experimental results show that the proposed method not only can improve the efficiency of image registration, but also can process higher registration accurately, and applicable for registering large-size remote sensing images.

关 键 词：大幅面遥感影像 SIFT 影像快速配准 地理信息约束 

分 类 号：P273.3[天文地球—测绘科学与技术]