各向异性扩散滤波的三维散乱点云平滑去噪算法

Study on Smooth Denoising of 3D Scattered Point Clouds with Anisotropic Diffusion Filtering

针对传统点云去噪算法在去除噪声时易造成模型特征失真的问题,提出一种各向异性扩散滤波的三维散乱点云平滑去噪算法.首先采用张量投票算法计算采样点的张量矩阵,并求解其特征值和特征向量;然后根据采样点的几何特征设计扩散张量的特征值,保证在不同特征方向的扩散速率能自适应调整;最后将重构的扩散张量与三维各向异性扩散滤波方程相结合,构造了点云滤波模型用于点云去噪.对不同含噪点云模型进行去噪的实验结果表明,该算法在点云去除噪声的同时,可以有效地保持原始模型的特征信息,避免了模型的过光顺.

To solve the distortion problem of model feature information during the denoising process based on the traditional point cloud denoising algorithm,a 3D scattered point cloud smoothing denoising algorithm was proposed based on anisotropic diffusion filtering.Firstly,the tensor matrix of the sampling point was obtained by tensor voting algorithm,and the eigenvalues and eigenvectors of the tensor matrix were solved;secondly,to adjust adaptively the diffusion rate of the different characteristic direction,the eigenvalues of diffusion tensor were designed based on the geometric characteristic of sampling point;finally,a point cloud filtering model was constructed for denoising by combining the reconstructed diffusion tensor with the 3D anisotropic diffusion filtering equation.The experimental results of different point cloud models with noise showed that the feature information of the original model was preserved effectively and the noise from the point cloud was removed.Therefore,the excessive smoothing of the model was avoided.