Robust registration of multi-modal remote sensing images based on multi-dimensional oriented self-similarity features

Yongjun Zhang, Wenfei Zhang, Yongxiang Yao, Zhi Zheng, Yi Wan, Mingtao Xiong

Abstract: Highlights•A lightweight multi-modal remote sensing image feature hybrid registration framework is proposed, which includes hybrid feature coarse matching and multi-dimensional oriented self-similarity feature fine matching.•While maintaining the geometric invariance of multi-modal remote sensing images, high-precision matching of corresponding points is achieved.•A representation method based on multi-dimensional oriented self-similarity features is proposed, which improves the utilization of features and promotes registration efficiency.•The effectiveness of the proposed algorithm is verified through experiments on multi-modal datasets.