目前双边滤波算法的空间方差σs和灰度方差σr凭借经验预先设定参数.然而,人工设置参数往往取值不合理,以致出现噪声不能滤除或者不同区域存在降噪和保留边缘特征之间矛盾的现象.文中提出一种自适应参数选择的双边滤波点云去噪算法,该算法首先为每个采样点建立k-邻域,然后以该点处的微切平面为视平面,在该视平面上利用目标尺度信息实现空间方差σs局部自适应取值,从而实现双边滤波算法自适应以保留更多的边缘特征.实验结果表明,所提出的方法滤波效果较传统双边滤波在边缘特征细节失真方面明显提高.%The existing technology can be only on bilateral filtering algorithm space variance σs and gray variance σr with the experience of the preset parameters.However,the artificial setting parameters are often unreasonable,so that the noise can not be filtered out or there is a contradiction between noise reduction and edge retention characteristics in different regions.This paper proposes an adaptive parameter selection of bilateral filtering point cloud denoising algorithm,the algorithm first for each sampling point set k-nighborhood,and then to the point of the tangent plane as the plane,to achieve local adaptive spatial variance σs value using the scale information of the target in the plane,so as to realize the adaptive bilateral filtering algorithm to preserve the edge feature.The experimental results show that the filtering effect of the proposed method is significantly improved compared with the traditional bilateral filtering in terms of edge feature detail distortion.
展开▼
