摘要
针对传统车载点云滤波方法在复杂地形条件下难以精确区分地面点、非地面点的问题,本文提出一种数学形态学滤波改进方法。该方法首先通过精细划分,将滤波区域分解为多个子区域,采用不规则三角网技术对每个子区域的地形坡度及其方法进行分析。之后,将子区域旋转至水平方向,以便最大化形态学滤波中开运算技术的性能,进而精确识别出点云数据中的非地面点,保证地面点数据的精确性。为了充分验证本文方法的实际表现,特别选取了包含平坦区域与斜坡区域的点云数据集进行测试。测试结果表明,本文方法在处理车载激光点云方面表现突出,尤其在坡度较大的区域,其滤波效果与性能均展现出了优越的水平。
To address the challenge faced by traditional vehicle-borne point cloud filtering methods in accurately distinguishing ground points from non-ground points under complex terrain conditions,this paper proposes an improved mathematical morphological filtering approach.This method first meticulously divides the filtering area into multiple sub-regions and employs a triangular irregular network(TIN)technique to analyze the terrain slope and its orientation in each sub-region.Subsequently,the sub-regions are rotated to a horizontal orientation,maximizing the effectiveness of the opening operation in morphological filtering.This enhancement facilitates the precise identification of non-ground points within the point cloud data,ensuring the accuracy of ground point data.To thoroughly verify the practical performance of the proposed method,point cloud datasets encompassing both flat and sloped regions were specifically selected for testing.The testing results demonstrate that the method exhibits outstanding performance in processing vehicle-borne laser point clouds,particularly in regions with steep slopes,where its filtering effects and overall performance are superior.
作者
圣桂锋
SHENG Guifeng(Zhaoqing Natural Resources Surveying and Mapping Institute,Zhaoqing 526060,China)
出处
《测绘与空间地理信息》
2025年第5期167-169,173,共4页
Geomatics & Spatial Information Technology
关键词
形态学改进滤波方法
车载激光点云
不规则三角网
开运算
improved morphological filtering method
vehicle-borne laser point clouds
triangular irregular network(TIN)
opening operation
作者简介
圣桂锋(1986-),男,安徽含山人,工程师,本科学历,主要从事测绘信息化工作。
