鲁棒的模糊C均值和点云双边滤波去噪

Robust Fuzzy C-Means and Bilateral Point Clouds Denoising

提出了一种将模糊C均值(FCM)聚类算法与双边滤波方法结合进行三维点云数据的去噪算法.该算法首先用模糊C均值聚类算法对大尺度噪声进行去除,并对小尺度噪声进行一定程度的光顺;然后用点云双边滤波器对小尺度噪声进行光顺.该算法将噪声分成大尺度和小尺度分步处理,不需迭代计算,提高了计算效率,也避免了光顺过程中产生过光顺问题.实验结果表明,本文方法可以在较好地保留尖锐特征的同时,有效地去除噪声.

A point clouds denoising method is presented which combines fuzzy c-means clustering with bilateral filtering approach. Firstly, large-scale noise, partly smooth small-scale noise is deleted with improved method of fuzzy c-means clustering. The cluster centers are regarded as the new points. Secondly, after acquiring new point sets being less noisy, the algorithm smoothes the remains noise by bi- lateral point clouds denoising method. The improved algorithm deals with the noise after separating it into large-scale and small-scale, which can avoid the iterative calculation, improve the computational efficiency and avoid the oversmoothing. The experimental results show that our new approach can reserve sharp feature and delete the noise effectively.