基于单光子探测的LiDAR三维点云数据处理方法设计

Design of LiDAR 3D point cloud data processing method based on single photon detection

针对移动曲面拟合法在处理三维点云数据时需要人为干预阈值的主要缺点,引入数学形态学的腐蚀运算,在处理采用TCSPC技术的机载激光测绘雷达获取的曲面目标物形成的三维点云数据的背景下,提出了"基于数学形态学改进的移动曲面拟合算法"。同时,为了提高数据利用效率,避免信息丢失,提出了"先去噪,再补点,后迭代"的去噪处理思想,在不降低数据密度的情况下,高效、稳健地进行噪声处理,在噪声水平为50%的情况下,总误差可以降至3.9%。

It introduce the corrosion of mathematical morphology operations, puts forward the“improved moving surface denoising algorithm based on mathematical morphology”,in view of the major drawback that the threshold of moving curved surface fitting need for human interventionwhen processing three-dimensional point cloud data. The mothed is suitable for the three-dimensional point cloud data of surface target accessing from LiDAR with TCSPC technique. At the same time, in order to improve the efficiency of data utilization and avoid the loss of information, a new denoising approach of “first to remove noise, then iteration after patch” is established. And without reducing the data density, the noise is processed efficiently and stably. Finally, through the experiment, the total error can be reduced to 3. 9% when the noise level is 50%.