结合间接邻域的散乱点云三角网格曲面重构

摘要

对于非均匀散乱点云，多数基于区域生长方法的曲面重构往往容易出现孔洞等缺陷。针对该问题，在K邻域点集的基础上提出间接邻域点集的概念，对以点为生长对象进行区域生长的三角网格曲面重构方法进行了研究，实现三角网格曲面重构。以生长点的邻域点集为样点估算微切平面，将邻域点投影至该平面上，并按照右手定则、逆时针方向进行排序，通过拓扑正确性原则从点列中去除错误的连接点，优化局部网格，选择较好的连接点，实现网格曲面的区域生长。%For uneven scattered point cloud, some defects such as holes may occur during surface reconstruction based on region growing method. To solve the problem, a novel concept of indirect neighborhood point set is proposed on the basis of K neighborhood point set. Combining indirect neighborhood point set with K neighborhood point set as neighborhood point set, a novel method of triangular mesh surface reconstruction using region growing through point is studied to realize triangular mesh surface reconstruction. It takes the neighborhood point set as the sample point to estimate micro tangent plane and project the neighborhood point set into this plane, ranks the points in counterclockwise direction according to the right-hand rule, removes the wrong points through checking the topological correctness of the local grid, optimizes the local grid, and then realizes triangular mesh surface reconstruction.