Improved Point Cloud Coarse Registration Algorithm And 3D Reconstruction Technology

With the development of computer technology applications and 3D scanning applications,point cloud 3D reconstruction technology is widely applied to various fields such as reverse engineering,medicine,digital preservation of cultural relics,and 3D animation.At the same time,point cloud registration technology occupies the dominant position of 3D point cloud reconstruction.Point cloud registration finds an optimal solution by calculating point cloud 3D data from different perspectives,maximizing local point cloud data and integrity.The 3D point cloud model is aligned.The main work of this paper is to improve the 4-Points Congruent Sets for Robust Pairwise Surface Registration.First,the initial point cloud is downsampled to obtain the initial point set and the target point set.In the initial point set,three points that are far apart but in the overlapping area are selected,and then the last point is selected,and the three points collinearity is eliminated and The invalid point in the triangle region consisting of three points;secondly,when the corresponding point is selected,the two invariants of the angle between the intersecting line segment and the four-point area are introduced,and the corresponding point pairs are classified;The number method solves the rotation translation matrix and selects the optimal solution according to the objective function to register the point cloud.For point clouds that have been coarsely matched,the point cloud is refined by the ICP(Iterative Closest Point)algorithm.At the same time,in order to better the 3D reconstruction of the point cloud in the later stage,this paper studies the 3D reconstruction technology of point cloud,adds the average grid method downsampling and bilateral filtering method to process the point cloud before the 3D reconstruction algorithm,and reconstructs the Power Crust.The algorithm and Poisson surface reconstruction algorithm are compared in terms of reconstruction effect.The point cloud 3D reconstruction technology flow in this paper includes improved point cloud coarse registration algorithm,ICP(Iterative Closest Point)fine registration algorithm,average grid method downsampling,bilateral filtering and Power Crust algorithm.The experimental results show that the improved point cloud coarse registration algorithm improves the registration accuracy and the registration effect is better than other coarse registration algorithms,which proves the effectiveness of the algorithm.At the same time,the 3D reconstruction technology flow is in the reconstruction point.The cloud model has outstanding effects.