三维重建中的点云精简研究-计算机应用技术专业论文.docxVIP

中北大学学位论文 中北大学学位论文 万方数据 万方数据 Research on Point Cloud Simplification of Three-dimensional Reconstruction Abstract In recent years, as the increase of peoples focus on the virtual reality technology, how to get a better three-dimensional reconstruction effect causes more researchers attention, and in the three-dimensional reconstruction process, the simplification of the three-dimensional point cloud data is the most important part of the three-dimensional point cloud data processing, so the research on point cloud simplification of three-dimensional reconstruction is a research hotspot. This paper mainly focuses on how to get a three-dimensional point cloud by the three-dimensional point cloud data of the different viewpoints and the three-dimensional simplification after pretreatment. The important parts of this paper are as follows: Firstly, by using the Kinect camera to obtain a three-dimensional point cloud data under the different viewpoints of the objects, when pretreatment, an improved ICP algorithm on different viewpoints was presented and uses this algorithm for three-dimensional point cloud registration under the different viewpoints to get the whole point cloud data of the objects. Secondly, an improved algorithm of the simplification of the three-dimensional point cloud data is described. By using octree for three-dimensional point cloud data topology division, and then use the method which combine the curvature and normal vector to get the point cloud feature information, according the information to simply the three-dimensional point cloud, which avoiding the excessive simplification of three-dimensional point cloud on the curved surface, and make better reconstruction results. Finally, for the features of Kinect camera in getting three-dimensional point cloud data, this paper present a simplification method which combine the color image, this method can avoid the feature information loss of three-dimensional point cloud data where the curvature of the s