DCPLD-Net: A diffusion coupled convolution neural network for real-time power transmission lines detection from UAV-Borne LiDAR data

Chi Chen, Ang Jin, Bisheng Yang, Ruiqi Ma, Shangzhe Sun, Zhiye Wang, Zeliang Zong, Fei Zhang

Abstract: Highlight•Propose DCPLD-Net for efficient extraction of powerlines from streaming LiDAR data.•A novel hybrid detector that combines simulated physical processes and neural network.•DCPLD-Net achieves an average F1 score of 97.14 % at 8 Hz detection frequency.