Research on pose estimation method for cooperative target based on monocular images

A monocular vision-based approach to spatial cooperative target pose solving

“…Yong Zhang et al [7] designed an infrared cooperative target, extracted the line features in the cooperative target, and used the principle of extinction lines to obtain the positional parameters of the UAV. Guo et al [8] used four non-coplanar lightemitting diodes as feature points and used the Posit algorithm to solve the relative positional attitude of the target in the single camera acquisition image. In the process of cooperative target-based camera relative pose solution, the feature point coordinates extraction from the image to be measured and the pose estimation algorithm are two key parts of the pose measurement system [9,10] In this paper, the matching point pairs in the image extract the feature point coordinates from the image to be measured, combine the dimensional information of the cooperative target and the camera calibration parameters, and solve the relative position and pose relationship between the camera coordinate system and the target coordinate system The The efficiency and accuracy of the feature point coordinate extraction have a great impact on the efficiency and accuracy of the pose solution.…”