Application Research of ship overload identification algorithm based on lidar point cloud

Abstract

Aiming at the problem that the structural characteristics of lidar 3D point cloud of dynamic ship are complex, which leads to the unsatisfactory segmentation of ship board, a dynamic ship overload identification algorithm based on reduced dimension segmentation is proposed in this paper. Different from the traditional three-dimensional point cloud edge extraction algorithm based on normal vector, this algorithm combines the unique point cloud structure characteristics of the ship and uses the dimension reduction segmentation method to accurately and quickly segment ship's board features on the projected two-dimensional point cloud. Then, through the memory relationship of the point cloud data, the two-dimensional segmentation results are regressed to the three-dimensional point cloud, a more reliable and accurate ship board point cloud can be obtained. The experimental results show that the algorithm can effectively improve the reliability and accuracy of ship board point cloud segmentation and ship overload identification when the structural characteristics of ship point cloud change dynamically.