点云曲面空间网格化加密求交算法

通过分析现有图形截交线、相贯线求解方法的优缺点，提出一种点云曲面空间网格化加密求交算法.采用几何图形离散化表达，并采用离散点求交集或重合度的方式计算图形间的公共部分.用空间网格包络盒快速定位点云曲面的相交区域，并采用计算三角面的重心位置，对相交区域进行点云加密.通过实际点云模型算例，验证该算法的有效性.经试验证明，所设计的算法操作简单、计算精度高、稳定可靠、适应性广.
Abstract: An intersection algorithm of point cloud surface by spatial mesh and refinement was proposed according to the analysis of the advantages and disadvantages of the existing methods of computing line of intersection. The discrete method was used to express the geometric graphic; intersection lines between the graphics were computed by intersection or coincidence degree of discrete points. The intersection region of the point cloud surface was quickly located by the space meshing envelope boxes. The point clouds in the intersection were refined by calculating the triangles' centre-of-gravity position. The validity of the proposed algorithm was verified by computing the intersection line in practical point cloud models. Results show that the proposed algorithm is simple, accurate, stable and reliable, with wide adaptability.