A Line Segment Detector for Space Target Images Robust to Complex Illumination

The relative pose estimation of the space target is indispensable for on-orbit autonomous service missions. Line segment detection an important step in estimation. traditional line detectors show impressive performance under sufficient illumination, while it easy to fail complex illumination conditions where too bright or dark. We propose a robust detector applications considering environments. An improved two-dimensional histogram construction strategy used optimize Otsu method improve accuracy anchor map extraction. To further detection’s effect, we introduce aggregation that uses angle difference between segments, distance endpoints, and overlap degree segments filter candidate connect disjoint probably came from same segment. demonstrate proposed using variety images collected on semiphysical simulation platform. results our has better than including LSD, Linelet, etc., terms precision.