航空学报 > 2007, Vol. 28 Issue (1): 161-166

基于点特征的干涉合成孔径雷达(InSAR)复图像自动配准算法

刘宝泉,冯大政,武楠,李军侠   

  1. 西安电子科技大学雷达信号处理重点实验室
  • 收稿日期:2005-08-08 修回日期:2006-04-18 出版日期:2007-01-10 发布日期:2007-01-10
  • 通讯作者: 刘宝泉

An image automatic registration method for InSAR complex images based on point features

LIU Bao-quan,FENG Da-zheng,WU Nan,LI Jun-xia   

  1. National Lab. for Radar Signal Processing, Xidian University
  • Received:2005-08-08 Revised:2006-04-18 Online:2007-01-10 Published:2007-01-10
  • Contact: LIU Bao-quan 

摘要:

在基于点特征的遥感图像配准过程中,特征点的自动提取和准确匹配是影响配准精度的关键。本文提出了一种干涉合成孔径雷达(InSAR)复图像对的自动配准算法,利用特征点的高曲率结构设计了模板,完成了特征点检测;根据匹配点对之间距离相近的结论设计了匹配算法,进行了特征点对的匹配。文章首先通过边缘检测和模板相关提取特征点;其次根据提出的匹配算法建立点的对应关系;最后利用两步法完成复图像的亚像元级配准。实验结果表明,该算法具有较高的配准精度。

关键词: 遥感信息工程, 图像配准, 两步法, 特征点匹配, 干涉合成孔径雷达, 亚像元级

Abstract:

Detecting feature points automatically and accurate feature matching are the two crucial steps in InSAR image registration. In this paper, an automatic feature-based registration algorithm is presented. We concentrate on two major steps: feature detection and feature matching. In feature detection, we design the templates to extract feature points by utilizing its high curvature property. In feature matching, we complete the process of the feature matching by using the matching algorithm proposed according to the conclusion of the near distance.

Key words: Remote , sensing , information , engineering, Image , registration, Two-step , method, Feature , points , matching, InSAR, Sub-pixel , accuracy

中图分类号: