Point Cloud Registration Method Based on SIFT Feature Points
ZHANG Shaoyu
Abstract：Point cloud registration is one of the key problems in the process of 3D modeling.Classical ICP algorithm is slower, especially in the case of large point cloud data takes longer time. Therefore an improved method is proposed. Firstly get the voxel grid sampling of the two pieces of point cloud,secondly find out the SIFT feature points of the source point cloud and save them into a piece of point cloud,and then call the ICP algorithm,using the saved SIFT feature point point cloud to regist the target point cloud,and remove the false matches through RANSAC algorithm,lastly keep the transformed source point cloud and repeat the above process until meet the convergence condition to improve the precision.Experiments show that the method is not only faster, but also more accurate.
Keywords：point cloud registrationICP algorithmSIFT feature pointsRANSAC algorithm
Publication Date：2018-01-01
Online Publishing Date：2025-08-15(First online date of this platform, not the publication date of the document)
Pages：5( 449-453 )