一种改进的多源遥感影像SIFT算法匹配策略

An improved SIFT algorithm matching strategy of multi-source remote sensing images

传统SIFT算法在遥感影像中相似性场景大量存在的情况下呈现出匹配效率低的问题。文中提出对传统SIFT算法匹配策略进行改进,其步骤包括:首先在初匹配阶段,利用建立金字塔影像匹配的方式确定影像间的几何关系模型;其次,利用该几何关系模型对初始影像进行约束匹配;最后利用RANSAC模型剔除误匹配点。通过相似性场景影像匹配实验可以表明,与传统SIFT算法相比,本方法具有明显的匹配效率和精度优势。

Traditional SIFT algorithm presents the problem of low matching efficiency in remote sensmg image with a lagrge number of similarity scenes. In order to solve this problem, this paper improves the matching strategy of SIFT algorithmthe. The steps include： in the initial matching stage, determine the geometric relation model between the images by matching based on establishing the pyramid images;then the geometric relation model is used to conteaint matching for the initial images; finally eliminate the mismatching points through RANSAC model. The images with similarity scene matching experiment results verify that this method has obvious efficiency advantage and matching precision compared with the traditional SIFT algorithm.