Power Line Extraction based on Feature Evaluation by Entropy Weight Method

Extracting power lines from point cloud data is important for power line monitoring applications. However, due to the high density of point cloud data acquired by UAV lidar and the complex scene of the power transmission pathway, it is still challenging to achieve high-precision and efficient power line extraction. To achieve high-precision electricity power line extraction, in this paper, we propose a method for power line extraction and reconstruction based on feature evaluation by the entropy weight method. The geometric distribution and covariance characteristics of the power line point cloud are analyzed, and different weight values are assigned to the features by calculating the weight information of each feature. Experimental results validate the accuracy of the proposed method for power line extraction.