基于二维Gamma分布的激光点云去噪

利用地面激光扫描仪测量时,由于空气中尘埃等因素会产生噪声点,对后期三维建模等应用带来不利影响。在现有研究成果基础上,深入分析散乱点云的近邻域特性,采用基于二维Gamma分布实现点云去噪,并在kd-tree索引支持下对其进行优化,通过邻域均值和邻域距离变化斜率两个约束条件共同移除噪声点。实验结果表明,本文算法能自动识别噪声点,降低人工设置阈值的影响,并与传统的基于正态分布的邻域点去噪算法进行对比,实验效果较优,达到了预期效果。

Laser point cloud removal based on 2D Gamma distribution

When measuring with a ground laser scanner,noise points are generated due to factors such as dust in the air,which adversely affects applications such as post-three-dimensional modeling.Based on the existing research results,the near-neighborhood characteristics of scattered point clouds are deeply analyzed,and the point cloud removal based on two-dimensional Gamma distribution is implemented and optimized by kd-tree index,through neighborhood mean and slope.The constraints together remove the noise points.The experimental results show that the proposed algorithm can automatically identify the noise points and reduce the impact of manually setting the threshold,and compare it with the commonly used neighborhood-based point denoising algorithm.The experimental results are better and the expected results are achieved.