基于空间子区一致性的异源图像匹配方法

摘要

提出一种基于空间子区一致性的异源图像匹配方法.该方法中,将图像划分为许多大小相同的子区;分别计算邻近子区间的灰度直方图距离,得到一种图像结构特征描述--似邻矩阵;利用似邻矩阵计算图像间的子区一致性;采用改进的搜索策略,快速获得匹配结果.实验结果表明,该方法在正确率和计算时间方面均优于传统方法.%A novel multi-sensor image matching method based on spatial sub-area congruency is proposed. In this method, images were segmented into sub-areas with same size. The stnucture feature represented by matrix of surrounding similarity was obtained by calculating the similarity of intensity histogram of each pair of adjacent sub-areas, and was used to compute spatial sub-area congruency.Through employing optimized searching algorithm, the matching result was acquired. Experimental results show that the proposed method can match multi-sensor images stably and fast and outperforms conventional methods.