首页 > , Vol. , Issue () : -
针对现有机载激光雷达(LiDAR)点云滤波算法难以准确分离复杂地形中地面点与非地面点问题,本文提出了一种基于点的多尺度形态学重建滤波方法(Point-based Multi-scale Morphological reconstruction Filter, PMMF)。在初始尺度层次下,PMMF主要通过构建一种基于点的形态学重建对原始点云进行滤波,即先在掩膜点云约束下借助k邻域结构元素和自适应高程缓冲区反复膨胀标记点云,获取潜在地面点；然后通过自适应坡度剔除潜在地面点中的非地面点,其中,坡度阈值可随地形复杂度自适应变化。在上层次尺度滤波结果基础上,PMMF提升种子点选择的网格尺度,重复上层次滤波过程,直至结果收敛。以国际摄影测量与遥感学会(ISPRS)发布的15组基准数据为研究对象,将PMMF滤波结果与近五年(2016-2020)提出的15种滤波算法比较。结果分析表明,PMMF有8组数据滤波效果占优,15组数据平均总误差和Kappa系数分别为2.71%和91.08%。使用四种不同地形特征的高密度机载LiDAR点云数据进一步验证PMMF的滤波效果,并将计算结果与渐进形态学滤波(PMF)、布料模拟滤波(CSF)、渐进加密三角网滤波(PTD)和多分辨率层次滤波(CSF)比较。结果表明,PMMF滤波性能最优,平均总误差为3.24%,较其它四种滤波方法分别减小了12.0%、59.1%、70.1%和53.2%。
Over the past decades, many airborne LiDAR point cloud filters have been proposed. However, these existing filters cannot obtain satisfactory results on complex landscapes, such as rugged slopes covered with low vegetation and discontinuous terrain. To overcome these problems, a Point-based Multi-scale Morphological reconstruction Filter (PMMF) is presented in this paper. Different from the classical morphological filters, PMMF takes raw point cloud rather than rasterized grids as the basic processing element. Firstly, potential ground points are obtained by repeatedly dilating the marker point cloud with the k-neighbor structural element and adaptive elevation buffer under the limits of the mask point cloud. Then, the non-ground points mixed in the potential ground points are eliminated by a terrain-adaptive slope filter. Based on the filtering results from the previous scale, PMMF increases the grid scale for selecting ground seeds, and repeats the filtering process as the previous level until the result converges. The three main contributions of the new algorithm include a point-based morphological method rather than a grid-based one to avoid information loss caused by point cloud rasterization, a multi-scale geodesic dilation with a slope-adaptive elevation buffer to select potential ground points and reduces the omission error on steep terrain, and a terrain-adaptive slope filter to eliminate commission errors mixed in potential ground points. PMMF was employed to filter the benchmark samples provided by ISPRS and its results were compared with 15 filtering algorithms proposed in the last five years (2016-2020). Results illustrate that PMMF outperforms the other filtering methods on 8 out of 15 samples, and its average total error and Kappa coefficient were 2.71% and 91.08%, respectively. Moreover, PMMF was used to process four high-density airborne LiDAR point clouds with different terrain features, and the filtering results were compared with progressive morphological filter (PMF), cloth simulation filter (CSF), progressive TIN densification (PTD) and multiresolution hierarchical filter (MHF). Results show that PMMF with an average total error of 3.24% has the best performance. Compared to PMF, CSF, PTD and MHF, the total error of PMMF is reduced 12.0%, 59.1%, 70.1% and 53.2%, respectively. A large number of experimental results show that PMMF has achieved satisfactory filtering results on various terrains, and the filtering accuracy is significantly higher than other conventional filtering algorithms. Experimental verification shows that the three innovations proposed in this paper are contributed to the higher accuracy of the new algorithm and overcome the imperfection of existing algorithms.