一种基于模糊C均值和均值滤波的点云去噪算法.pdfVIP

·机械制造 · 许龙,等 ·一种基于模糊C均值和均值滤波的点云去噪算法 一种基于模糊 C均值和均值滤波的点云去噪算法 许龙,黄翔,李根 (南京航空航天大学 机电学院,江苏 南京210016) 摘　 要:因点云数据中存在噪声,通常对不同特征的点云数据采用相同的处理方法,虽然能删 除噪声但也会因删除尖锐特征造成过光顺。 提出了一种基于模糊C均值(FCM)聚类算法且均 值滤波的点云去噪算法。 该算法使用模糊C均值聚类算法删除大尺度噪声后,再将均值滤波 应用到点云光顺中,对数据点中的小尺度噪声进行光顺。 实验结果表明,该算法去噪效果明 显,在去噪光顺过程中较好地保持了边界特征,也避免了过光顺问题的产生。 关键词:模糊C均值;点云;均值滤波;去噪 中图分类号:TH123　 　 文献标志码:B　 　 文章编号:1671-5276(2016)04-0005-03 - - Point Clouds De noise Algorithm Based on Fuzzy C means and Mean Filter XU Long,HUANGXiang,LI Gen (College of Mechanicaland Electronic Engineering,Nanjing University of Aeronautics and Astronautics,Nanjing210016,China) Abstract:Because noise exists in point cloud algorithm,the different treating methods are adopted in the algorithms. Though this - - method can be used todenoise,over smoothingcoursed by generalde nosingalgorithmiscut offtoo.This paperpresentsathree - - - - - dimensionpointcloudsde noisingalgorithmcombinedwithfuzzy c means(FCM) ,meanfiltering.Thelarge scalenoiseiscutoff - - - by the improvedfuzzy c means andthesmall scale noise is smoothed by the improved meanfilter. Experimental results showthat - - this approachcanbe usedto reserveboundaryfeatureandeffectively de noise.Meanwhile,theproblemofover smoothingof point cloud is effectively avoided. Keywords: - - - fuzzy c means;point clouds;meanfiltering;de nosing