改进贪婪投影三角化算法的激光点云快速三维重建

Fast 3D reconstruction of point cloud based on improved greedy projection triangulation algorithm

通过三维激光扫描仪获取的点云数据具有密度大、精度高等特点。本文针对贪婪投影三角化算法在对采集的大量点云数据进行三维重建时耗时长,重构的模型表面不够光滑,存在细小孔洞的问题,提出一种改进的点云三维重建算法。该方法首先用体像素网格滤波算法对点云进行下采样;然后使用移动最小二乘算法对输入的点云进行平滑及重采样,并且使用八叉树来代替KD树进行近邻域搜索;最后使用基于移动最小二乘算法的点云法线估计的贪婪投影三角化算法对点云进行重建。经过实验验证,该方法可以缩短重建时间,减少孔洞,并构建出平滑、点云拓扑结构更为准确的模型。

The point cloud data obtained by 3D laser scanner has the characteristics of high density and high accuracy.Aiming at the problem that the greedy projection triangulation algorithm takes a long time to reconstruct a large number of point cloud data,the reconstructed model surface is not smooth enough,and there are small holes,thus an improved point cloud 3D reconstruction algorithm is proposed.Firstly,the point cloud is downsampled using a voxel grid filtering algorithm.Then,the moving least square algorithm is used to smooth and resample the input point cloud,and the octree is used to replace the KD tree for neighborhood search.Finally,the greedy projection triangulation algorithm based on moving least squares algorithm is used to reconstruct the point cloud.The experimental results show that the proposed method can shorten the reconstruction time,reduce the number of holes,and build a smooth and more accurate point cloud topology model.