Sieving Scattered 3D Point Clouds Using Clustering Analysis for 3D Surface Reconstruction

During the course of 3d surface reconstruction,there are a large number of noises and isolated 3d points in raw 3d point clouds,which obtained from images.If we directly use these data to reconstruct surface,the algorithm will make surface sharply prominent and ineffective reconstruction.Because of above problems,a method that sieving 3d point clouds based on DBSCAN is presented in this paper,and then 3d surface is reconstructed using filtered 3d point clouds.Experiments show that good 3d surface reconstruction is obtained using this algorithm.