Abstract:
The detection and extraction of individual pylons and power lines from high-density point cloud (PC) LiDAR data are a relevant tool for evaluating the power lines utility...Show MoreMetadata
Abstract:
The detection and extraction of individual pylons and power lines from high-density point cloud (PC) LiDAR data are a relevant tool for evaluating the power lines utility corridors. Moreover, the presence of high vegetation and hilly terrain is a research challenger in the available methods. The paper presents a novel method for the extraction of pylons and power lines. Two steps compose the proposed approach: a pylon detection step based on top view projection, denoted by DFSS - Detect Filled Square Shapes, and a pylon arms detection step with the DPA - Detect Pylon Arm algorithm. The results show that the proposed method could accurately and automatically extract pylons and the associated power lines, even if the dataset has low quality with downsampling, to reduce the processing time. Field tests were performed with a ground static LiDAR and a point cloud affected by downsampling voxel grid and Gaussian noise to simulate the expected LiDAR data from a UAV.
Published in: 2021 IEEE International Conference on Autonomous Robot Systems and Competitions (ICARSC)
Date of Conference: 28-29 April 2021
Date Added to IEEE Xplore: 18 May 2021
ISBN Information: