基于脉冲耦合神经网络的点云曲面去噪

PCNN-Based Point Set Surface Denoising

提出一种基于脉冲耦合神经网络(PCNN)的点云曲面去噪算法.该算法主要分为两步:噪声点定位和噪声点滤波.首先针对点云曲面构建一个PCNN神经网络,各个神经元的外部刺激值由邻近点的几何位置差异和法向差异构成,利用神经元输出的自适应点火捕获特性,实现了噪声点的定位;而后针对点云曲面中的噪声点,基于网格光顺中双边滤波的思想,实现噪声点的滤波,对于非噪声点,则保持原有的几何位置不变.实验结果表明,由于区分了噪声点和非噪声点,该算法较传统的点云曲面去噪算法能更加有效的去除噪声的同时并保持模型的几何特征.

A novel algorithm of PCNN-based point set surface denoising is proposed in this paper.The algorithm mainly includes two steps：location of noise points and smoothing of the located noise points.Firstly,a pulse-coupled neural network for the point set surface is constructed.The stimulation value of each neuron is decided by the differences of the position and the normal of the k-nearest neighbor points.The noise points are located through the adaptive firing capture feature of the PCNN.Based on the idea of bilateral filtering,the located noise points are smoothed,while the non-noise points remain their geometry position.Due to the different operations on noise points and non-noise points,experiments show that our algorithm performs better to remove the noise of the point set surface while keeping the features of the model.