摘要
针对三维扫描仪获取的含噪点云数据会严重影响到后期三维重建的精度,提出一种新的散乱点云快速去噪算法。该算法首先通过改进的K-means聚类算法来建立点云的空间拓扑关系,然后对聚类后每一类的点云进行噪声点识别及去除。实验结果表明算法简单快速,在散乱点云实现有效聚类的基础上不但去噪效果良好,而且能够快速去除点云中的明显离群噪声点,保留理想目标点云。
Noised point cloud data captured by 3D scanner can seriously affect the precision of three-dimensional reconstruction in late stage. In light of this, we present a new denoising algorithm for scattered point clouds. The algorithm first establishes the spatial topology relation of point cloud by the improved K-means clustering algorithm. Then it recognises and removes the noise points on every kind of clustered point cloud. Experimental results show that the algorithm is simple and fast. On the basis of effective clustering by scattered point cloud, the algorithm is not only good in denoising, but can also quickly remove the obvious outlier noises in point cloud and keeps ideal target point cloud.
出处
《计算机应用与软件》
CSCD
2015年第7期74-78,共5页
Computer Applications and Software
基金
重庆市教委项目(KJ100821)
重庆理工大学研究生创新基金项目(YCX2013218)
关键词
散乱点云
K-MEANS聚类算法
噪声点
去噪
Scattered point cloud K-means clustering algorithm Noise points Denoising
