Abstract:In the process of point cloud registration,the iterative closest point (ICP) algorithm is highly dependent on the initial position of point cloud and the iteration speed is slow.To solve the problem,a point cloud registration method based on scale invariant feature transform (SIFT) feature points combined with ICP is proposed.First of all,the SIFT algorithm is used to extract the feature points of the point cloud to be registered and the target point cloud; and the fast point feature histogram (FPFH) features of the feature points are calculated; afterwards,according to the feature and using sample consensus initial alignment (SAC-IA) algorithm,the initial transformation matrix is found,thus completing the initial registration; finally,on the basis of the initial registration,the ICP algorithm is used to precisely register the two point clouds.Experiments show that compared with the ICP algorithm,this method has better registration accuracy,while the efficiency is also significantly improved.