Research and Application of Automatic Classification Technology of LiDAR Point Cloud Data of Multi Rotor UAV for Transmission Line

• Main Authors: RUAN Jun, TAO Xiongjun, LIU Dongjia, ZHANG Chen
• Format: Article
• Language: English
• Published: Energy Observer Magazine Co., Ltd. 2019-06-01
• Series: 南方能源建设
• Subjects: airborne lidar; uav; danger analysis of tree barriers; point cloud automatic classification
• Online Access: https://www.energychina.press/en/article/doi/10.16516/j.gedi.issn2095-8676.2019.02.016

[Introduction] In order to improve the efficiency and quality of laser point cloud data processing and provide data guarantee for tree barrier detection for transmission lines. [Method] A new automatic point cloud classification algorithm was proposed based on the self developed laser radar UAV, then a tree barrier analysis software was developed based on the algorithm to realize one-click automatic classification of point cloud. [Result] The results show that the accuracy of point cloud classification can reach more than 95%, and the efficiency of point cloud automatic classification can reach 60 km/h, moreover, good classification results are achieved in the complex areas where the towers and power lines, towers and vegetation intersect each other. [Conclusion] This method provides a high efficiency, high quality, automatic and intelligent data processing method for the analysis of tree barrier danger, and effectively improves the acquisition accuracy and processing efficiency of geospatial three-dimensional information data.