The International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences
Download
Publications Copernicus
Download
Citation
Articles | Volume XLVI-3/W1-2022
Int. Arch. Photogramm. Remote Sens. Spatial Inf. Sci., XLVI-3/W1-2022, 141–148, 2022
https://doi.org/10.5194/isprs-archives-XLVI-3-W1-2022-141-2022
Int. Arch. Photogramm. Remote Sens. Spatial Inf. Sci., XLVI-3/W1-2022, 141–148, 2022
https://doi.org/10.5194/isprs-archives-XLVI-3-W1-2022-141-2022
 
22 Apr 2022
22 Apr 2022

A 3D LIDAR RECONSTRUCTION APPROACH FOR VEGETATION DETECTION IN POWER TRANSMISSION NETWORKS

Y. Ma1, F. Zhou1, G. Wen1, H. Gen1, R. Huang1, Q. Wu2, and L. Pei2 Y. Ma et al.
  • 1Electric Power Research Institute, Yunnan Power Grid Company ltd., Kunming, China
  • 2School of Electronic Information and Electrical Engineering, Shanghai Jiao Tong University, Shanghai, China

Keywords: Vegetation Management, Tree detection, Deep Learning, Ground Plane, Localization, 3D Reconstruction

Abstract. Vegetation management is important to the power transmission and distribution networks. The encompassed towering tree is always the key factor of the high impedance faults(HIFs).LiDAR is an efficient way to detect trees with 3D point cloud. The classical tree detection algorithm can handle the tree with high and distinct trunk,but limited to the tree with messy trunks. While the deeplearning based detection algorithms are also suffered from the terrain noise points. In this paper, we propose an efficient LiDAR reconstruction system which can efficiently reconstruct the point cloud of surrounding vegetation without the ground plane noise. We also use different weight strategies to improve the localization accuracy. We have conducted our system on the real power network environment and the height detection result shows that our algorithm has a better accuracy and robustness compared with the classical methods.