计算机应用与软件2016，Vol.33Issue(10)：148-152,5.DOI:10.3969/j.issn.1000-386x.2016.10.033
一种聚类与滤波融合的点云去噪平滑方法
A POINT CLOUD DENOISING AND SMOOTHING METHOD BASED ON FUSION OF CLUSTERING AND FILTERING
摘要
Abstract
The original three-dimensional point cloud data collected has the problems of noise and unsmooth surface which is not conductive to three-dimensional post-reconstruction.In view of this,this paper presents a three-dimensional point cloud denoising and smoothing method,it is based on the fusion of adaptive density clustering algorithm and bilateral filtering.First,the method applies adaptive density clustering analysis on the point cloud model,and erases the noise points in the model according to clustering result.Then,it calculates the k neighbourhood of sampling point,and calculates the normal vector of the plane where the k neighbourhood is used to construct sampling points,and further obtains the bilateral filtering factor so as to smooth the point cloud model.Experimental results show that the proposed algorithm can identify and remove noise effectively and smoothes the point cloud model.At the same time,it can well keep characteristic information of the original model.关键词
点云去噪/自适应密度聚类/k邻域/双边滤波/特征保持Key words
Point cloud denoising/Adaptive density clustering/K neighbourhood/Bilateral filtering/Feature preserving分类
信息技术与安全科学引用本文复制引用
牛晓静,王美丽,何东健..一种聚类与滤波融合的点云去噪平滑方法[J].计算机应用与软件,2016,33(10):148-152,5.基金项目
国家高技术研究发展计划项目（2013AA 102304）；第56批中国博士后科学基金项目（2014M562457）。 （）
