Three Dimensional Point Cloud Processing Research

With the development of 3-D scanning techniques, how to process the three dimensional point cloud becomes one of the key research fields. This research thesis is from the Project of Hi-Tech Research and Development Program of China (863): Close object 3D measuring technique.In this paper, the author aims at the processing of spatial scattered point cloud. It includes several parts: data compression, data smoothing, neighbor finding, triangulation and triangle mesh reconstruction. The paper pays emphasis on data compression method of point cloud. By analyzing the feather of the point cloud, an octree model is established. An improved encoding algorithm of linear octree is presented. This method increases the continuity of the Morton code, decreases the depth of the octree, and improves data compression ratio. The tests are made to confirm the validity of the method. Furthermore, two nearest neighbor finding algorithms are introduced and compared in chapter 3. Triangulation and triangle mesh reconstruction on the scattered point are introduced in next part of the paper. Delaunay Triangulation and, five optimized principle and Hoppe's triangle mesh reconstruction algorithm are presented.