Optimal selection method of redundant orbiter images in planetary mapping based on multi-constraints

Abstract

With the increasing number of planetary orbiter images, how to quickly select optimal images from the redundant data for planetary mapping tasks has become a critical issue. This paper proposes a planetary orbiter image selection method based on multi-constraints. In this method, orbiter image metadata and various constraints in photogrammetry mapping are combined, and a subset of images that satisfies mapping requirements from a large amount of redundant orbiter data is obtained through three parts: metadata-based selection in the preprocessing, images selection before feature matching, and stereo image pairs selection before dense matching. The experimental results indicate that the mean elevation deviation between the Digital Elevation Model (DEM) generated by the selected data subset and by the raw data is less than 1 meter, the DEM has 100% coverage ratio in the region of interest, and the data redundancy ratio is reduced by more than 500%. This method improves mapping efficiency while ensuring mapping accuracy.