机载激光雷达(LiDAR)点云电力线提取过程中存在杆塔形状复杂、噪声影响大等问题,导致电力线点云提取精度低,本文提出一种基于点云分块处理、格网划分的曲面拟合滤波、自适应密度聚类算法的电力线点云提取与重建方法.首先,根据电力线走向,对整体点云进行分块处理;其次,在曲面拟合算法的基础上,引入格网划分思想,提出一种改进曲面拟合滤波算法并进行点云滤波;最后,通过给出自适应密度聚类解决方案精确提取电力线点云.借助点云库(PCL)、libLAS库与Visual Studio 2017 C++开发环境实现本文算法,基于实测点云数据对本文方法进行测试与精度评定.结果表明:电力线提取精确率为97.82%、召回率为99.76%、F1值为98.78%,一次便可实现电力线的成功提取,在保证提取精度的同时提升了提取效率,本文研究能够为电力线智能巡检提供良好的工程应用价值.
Power line extraction method based on unmanned aerial vehicle LiDAR point cloud data
In the power line extraction process of airborne light detection and ranging(LiDAR)point cloud,there are problems such as complex tower shape and great noise influence,which lead to low accuracy of power line point cloud extraction.In this paper,a power line point cloud extraction and reconstruction method based on point cloud block processing,surface fitting filtering with grid division,and adaptive density clustering algorithm was proposed.Firstly,according to the power line trend,the whole point cloud was segmented.Secondly,based on the surface fitting algorithm,an improved surface fitting filtering algorithm was proposed for point cloud filtering by introducing the idea of grid division.Finally,the power line point cloud was accurately extracted by the adaptive density clustering solution.Point cloud library(PCL),libLAS library,and Visual Studio 2017 C++development environment were used to implement the algorithm in this paper.The method was tested,and its accuracy was assessed by measured point cloud data.The results show that the accuracy rate of power line extraction is 97.82%;the recall rate is 99.76%,and the F1 value is 98.78%.The successful extraction of the power line can be realized in one time,and the extraction efficiency is improved while the extraction accuracy is guaranteed.The research in this paper can provide good engineering application value for intelligent power line patrol.
airborne light detection and ranging(LiDAR)point cloudpower line extractionimproved filtering algorithmadaptive density clusteringaccuracy evaluation
万方数据
维普
