Feature matching research of portal crane based on improved QB-SHOT feature descriptor

Aiming at the problems of missing information and low matching rate when using binary azimuth histogram (B-SHOT) for point cloud feature extraction and matching under the complex working conditions of port portal cranes, a B-SHOT feature descriptor (QB-SHOT) based on interval double coding bits is used to carry out feature extraction and feature matching to establish the correspondence between adjacent point clouds. Experimental results show that the improved QB-SHOT algorithm improves the algorithm speed and matching accuracy, and the experimental verification of the feasibility and effectiveness of QB-SHOT application in port portal crane.