本文已被：浏览 1829次 下载 5988次
Received:September 03, 2010 Revised:October 20, 2010
Received:September 03, 2010 Revised:October 20, 2010
中文摘要: 三维重建过程中获得的初始海量数据存在大量的噪声和孤立点，使得直接使用这些数据进行网格重建时，将会产生尖锐的凸出，导致重建效果不好，甚至是网格重建失败。针对以上问题，提出首先采用基于密度聚类的方法筛选三维点云，然后进行网格重建。实验表明本文算法获得了较好的网格重建效果。
Abstract:During the course of 3d surface reconstruction, there are a large number of noises and isolated 3d points in raw 3d point clouds, which obtained from images. If we directly use these data to reconstruct surface, the algorithm will make surface sharply prominent and ineffective reconstruction. Because of above problems, a method that sieving 3d point clouds based on DBSCAN is presented in this paper, and then 3d surface is reconstructed using filtered 3d point clouds. Experiments show that good 3d surface reconstruction is obtained using this algorithm.
文章编号： 中图分类号： 文献标志码：
基金项目:国家教育部春晖计划重点资助项目(Z2004-1-520);贵州省省长基金(黔省专合字(2007)14 号);贵州大学研究生创新基金(校研理工2010034)
引用文本：
陈晓霞,陈孝威.三维重建中散乱点云的聚类筛选与网格重建.计算机系统应用,2011,20(4):141-144
CHEN Xiao-Xia,CHEN Xiao-Wei.Sieving Scattered 3D Point Clouds Using Clustering Analysis for 3D Surface Reconstruction.COMPUTER SYSTEMS APPLICATIONS,2011,20(4):141-144
陈晓霞,陈孝威.三维重建中散乱点云的聚类筛选与网格重建.计算机系统应用,2011,20(4):141-144
CHEN Xiao-Xia,CHEN Xiao-Wei.Sieving Scattered 3D Point Clouds Using Clustering Analysis for 3D Surface Reconstruction.COMPUTER SYSTEMS APPLICATIONS,2011,20(4):141-144
