一种综合利用像方和物方信息的多影像匹配方法  

A Novel Method of Multi-image Matching Using Image and Space Synthesis Information

作　　者：袁修孝[1,2] 明洋[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

出　　处：《测绘学报》2009年第3期216-222,共7页Acta Geodaetica et Cartographica Sinica

基　　金：国家自然科学基金(40771176);国家创新研究群体科学基金(40721001)

摘　　要：提出一种综合利用像方和物方信息的多影像匹配方法。首先对各原始影像采用3×3像元平均法生成4层金字塔影像,并在最高层金字塔影像中计算各搜索影像与基准影像的初始视差;然后从第3层金字塔影像开始进行基准影像与各搜索影像的匹配,并对各立体像对的匹配结果进行物方融合,剔除部分误匹配点,获得较精确的物方空间信息,以用于下一层金字塔影像的匹配。此过程融入带几何约束条件的相关系数法匹配和整体松弛法匹配策略;最后通过对某地区一组大重叠度数码航空影像的匹配试验,证实该方法的有效性。通过检测所生成的DSM精度,表明相对于传统的双片影像匹配方法,多影像的匹配可以有效剔除误匹配点并显著提高影像匹配的成功率。A novel method is introduced for multi-image matching by synthesizing image and space information. Four level pyramid images are generated firstly according to the rule that the next level pyramid is generated from the previous level using the average gray values of the3×3 pixels, and the first level pyramid is generated from the original image. The initial horizontal parallaxes between the reference image and each searching image are calculated in the highest pyramid level. Then corresponding image points are searched in each stereo pairs from the third level pyramid images, and the matching results in all stereo pairs are integrated into the object space, by which the mismatched image points can be eliminated and more accurate spatial information can be obtained for the subsequent pyramid image matching. The method based on correlation coefficient with geometric constraints and global relaxation optimization is introduced in the process of matching. Finally, the feasibility of the method proposed in this paper is verified by the experiments using a set of actual digital aerial images with big overlap. Compared with the traditional matching with two images, the accuracy of DSM generated using our method shows that multi-image matching can eliminate the mismatched points more effectively and improve the success ratio of matching more significantly.

关 键 词：多影像匹配 数字表面模型(DSM) 带几何约束条件的相关系数法匹配 松弛法匹配 精度 

分 类 号：P231[天文地球—摄影测量与遥感]