复杂地形数据缺失下电力线点云自动提取方法

Automatic Extraction Method of Power Line Point Cloud Under Complex Terrain Data Missing

该文针对采用机载LiDAR扫描点云数据在复杂地形数据缺失下,提取电力线潜在区域不理想的情况,提出一种新的电力线点云自动提取方法。首先将点云数据转为栅格数据,利用Hough变换提取包含电力线点云的潜在区域;然后,利用主成分分析法,根据局部线型特征提取电力线种子点,依据种子点采用密度聚类算法,提取单条电力线种子点;最后基于电力线模型精确提取电力线点云。通过实例分析,结果表明:在复杂地形数据缺失的情况下,该方法提取率达94.83%,平均拟合残差为0.047m,得到了理想的提取结果。

In this paper,according to the condition that the extraction of potential area of power line is not ideal,through scanning point cloud data by airborne LiDAR when the complex terrain data is missing,a new automatic extraction method of power line point cloud is proposed.Firstly,the point cloud data is converted into grid data,and the potential area containing power line point cloud is extracted by Hough transform;Then,the power line seed points are extracted by using principal component analysis according to the local line features,and the seed points of a single power line are extracted by density clustering algorithm according to the seed points;Finally,the power line point cloud is accurately extracted based on the power line model.Through the analysis of an example,the results show that:In the case of missing complex terrain data,the extraction rate of this method is 94.83%,the average fitting residual is 0.047 m,and the ideal extraction results are obtained.