Point Cloud Registration Method for Pipeline Workpieces Based On NDT and Improved ICP Algorithms

Abstract

A new point cloud registration method based on NDT and improved ICP algorithm is proposed to solve the problem of point cloud registration of laser scanning workpiece position and pose data on industrial pipeline. Firstly, according to the FPFH algorithm, the feature points of point clouds are extracted to reduce the number of point clouds. Then, the normal distribution transformation (NDT) is used to achieve rough registration of point clouds, so that the two point clouds have relatively good initial position and posture. Finally, based on the traditional ICP algorithm, the k-d tree is used to accelerate the search speed of the corresponding point pairs and complete the accurate registration of the point clouds. The proposed algorithm and the NDT+ICP algorithm are tested respectively, and the results show that the proposed algorithm is faster, with lower error and fewer iterations, which provides a reference for similar problems.