Point cloud registration is the key step of 3D non-contact precision measurement for complex mechanical parts based on machine vision.Due to extreme dependence on initial position,slow iterative convergence speed and many wrong corresponding point pairs,iterative closest point (ICP) algorithm could not satisfy the requirements of point cloud registration efficiency and measurement precision of large quantities of complex mechanical parts,so an improved ICP method of measurement point cloud registration for complex mechanical parts by combining ICP with intrinsic shape signature-fast point feature histogram (ISS-FPFH) feature is proposed.In order to reduce the registration quantity of the point cloud and keep the original subtle features on surface of point cloud,the voxel filter based on the point close to center of gravity is proposed to preprocess for point cloud downsampling.It is difficult for traditional ICP algorithm to determine the appropriate initial position and will lead to the registration failure of multi-view measurement point cloud, the sample consensus initial alignment (SAC-IA) algorithm based on ISS-FPFH is applied for coarse registration.In order to solve the problems of slow iterative convergence speed and many wrong corresponding point pairs of traditional ICP algorithm,the point-to-plane ICP algorithm by combining with normal vector angle constraint is proposed for fine registration.The robustness of the proposed method to noisy point clouds is verified with the bunny point cloud model from Stanford University.The proposed method, traditional ICP algorithm and SAC-IA+ICP algorithm are tested and compared by taking the common complex mechanical blade and door handle as experimental objects.The results show that the root mean square error (RMSE) and registration time in the point cloud registration experiment for two different kinds of mechanical parts has an average reduction of 80.46% and 49.07% compared with the traditional ICP algorithm and an average reduction of 67.86% and 16.97% compared with the SAC-IA+ICP algorithm.Therefore,the proposed method can meet requirements of 3D non-contact precision measurement for large quantities of complex mechanical parts.