一种基于结构特征边缘的多传感器图像配准方法  

A Multi-sensor Image Registration Algorithm Based on Structure Feature Edges

作　　者：苏娟[1,2] 林行刚[2] 刘代志[1] 

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

出　　处：《自动化学报》2009年第3期251-257,共7页Acta Automatica Sinica

基　　金：国家自然科学基金(60472028)资助~~

摘　　要：图像配准是多传感器图像融合等处理的前提.本文以包含人造目标的合成孔径雷达(Synthetic aperture radar,SAR)图像和可见光图像为处理对象,提出了一种基于结构特征边缘的多传感器图像配准方法.该方法提取人造目标在两类图像中表现的共性特征——结构特征边缘,并基于边缘匹配构造虚拟角点,采用基于特征一致的粗配准方法和基于虚拟角点的精配准方法,对待配准图像实现由粗到精的自动配准.实验结果表明,本文方法能够取得较高的配准精度.Image registration is prerequisite to multi-sensor image processing, such as image fusion. Taking synthetic aperture radar （SAR） images and optical images containing man-made objects as examples, this paper proposes a multisensor image registration algorithm based on structure feature edges. The common features for the same man-made objects shown in the two kinds of images, i.e. structure feature edges, are extracted firstly, and virtual corners are constructed on the basis of edge matching. The coarse registration based on feature consensus and the fine registration based on virtual corners are used to realize the automatic coarse-to-fine registration between multi-sensor images. Experimental results show that the proposed algorithm has a high registration accuracy.

关 键 词：图像配准 结构特征边缘 边缘匹配 特征一致 虚拟角点 

分 类 号：TP391.41[自动化与计算机技术—计算机应用技术]