点云模型的谱聚类分割  

Point Cloud Segmentation Based on Spectral Clustering

作　　者：马腾[1] 龙翔[1] 冯路[1] 骆沛[1] 吴壮志[1] 

机构地区：[1]北京航空航天大学计算机学院,北京100191

出　　处：《计算机辅助设计与图形学学报》2012年第12期1549-1558,共10页Journal of Computer-Aided Design & Computer Graphics

基　　金：国家科技支撑计划(2008BAH29B00)

摘　　要：为了实现点云模型的有意义分割,提出一种基于谱聚类的分割算法.首先用图G表示点云模型,将分割问题转化为图切割问题;然后根据归一化的非对称Laplacian矩阵构造谱聚类空间;最后通过移除掉多余的特征向量,在一个更低维的空间中找到了分割问题的松弛解.文中还给出了该算法相关定理的证明,并通过实验验证了算法的正确性和有效性.A spectral clustering based method is proposed to segment point cloud into meaningful subparts. By representing the point cloud as a graph G, the segmentation problem can be turned into a graph miwcut problem. The nonsymmetric normalized Laplacian matrix is used to construct the spectral space. By removing redundant eigenvectors from the spectral domain, the segmentation solution is found in a lower dimensional space. The theoretical guarantee of the proposed method is proved. The accuracy and efficiency of the algorithm are verified by experimental results.

关 键 词：点云模型 点云分割 谱聚类 LAPLACIAN矩阵 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]