Research and Application of Automatic Classification Method for Patrol Targets of Transmission Lines Based on LiDAR Point Cloud
The development of airborne laser radar technology and cloud data processing technology provides a new technical means for the acquisition and processing of geospatial 3D data. Based on the self-developed LiDAR UAV, this paper proposes a new point cloud automatic classification method (TL-PCACM) of transmission lines by using the improved fast 3D convex hull construction algorithm, k-means clustering analysis algorithm and Region-growing algorithm, which have high classification accuracy, good classification effect and fast processing speed. The accuracy of point cloud automatic classification is over 95%, the efficiency can reach 60 km/h. The proposed method is applied to the analysis of tree obstacles under transmission lines. The efficiency of the automatic identification of tree obstacles can reach 720 towers/hour. The accuracy can reach 50 cm. It provides high efficiency, high quality, fully automatic and intelligent data processing method for the classification of tree obstacles, effectively improving the processing accuracy and speed of geospatial 3D information data.
KeywordsAirborne laser radar UAV Transmission line Point cloud automatic classification Analysis of tree obstacles
The research is supported by Implementation of Innovation Driven Development Capacity Building Special Funds of Guangdong Academy of Sciences (2017GDASCX-0101, 2017GDASCX-0601), Guangdong Innovative and Entrepreneurial Research Team Program (2016ZT06D336), Guangzhou Science and Technology Program (201604016047, 201806010106), Guangdong Science and Technology Program (2017B010117008), Guangzhou Hydraulic Technological Innovation Project (MZSK-2016-01, SW-2018-01), National Natural Science Foundation of China (41401430), Key Lab of Guangdong for Utilization of Remote Sensing and Geographical Information System (2017B030314138).
- 1.Xin, Z.: The “13th Five-Year” power network planning of China Southern Power Grid was fully launched. Guangxi Electr. Power (7), 12 (2013)Google Scholar
- 2.Duan, M.: 3D power line reconstruction from airborne LIDAR point cloud. Master’s thesis of Wuhan University, Hubei (2015)Google Scholar
- 4.Zhang, J., Duan, M., Lin, X., et al.: Comparison and analysis of models for 3D power line reconstruction using LiDAR point cloud. J. Wuhan Univ. Inf. Sci. Ed. 42(11), 1565–1572 (2017)Google Scholar
- 5.Brodua, N., Laguea, D.: 3D terrestrial lidar data classification of complex natural scenes using a multi-scale dimensionality criterion: applications in geomorphology. ISPRS J. Photogramm. Remote. Sens. Elsevier 68, 121–134 (2012). Egu General Assembly Conference, 2012, 14 (3): 4368CrossRefGoogle Scholar
- 6.Zhang, L.: The Method of Airborne Laser Scanning Data Classification. Southwest Jiaotong University, Chengdu (2013)Google Scholar
- 7.Yu, L., Li, T., Zhan, Q., et al.: Segmentation of Li-DAR point clouds based on similarity measure in multi-dimensional Euclidean space. Remote. Sens. Land Resour. 26(3), 31–36 (2014)Google Scholar
- 9.Wang, J., Li, C., Zhou, M.: Analysis of airborne full-waveform Li DAR data for supervised point cloud classification. Remote. Sens. Inf. 28(5), 21–27 (2013)Google Scholar
- 10.Fan, S., Zhang, A., Hu, S., et al.: A method of classification for airborne full waveform Li DAR data based on random forest. Chin. J. Lasers 9, 216–222 (2013)Google Scholar
- 11.Li, Z.: Study and Improvement of Fast Three-Dimensional Convex Hull Algorithm. Shanghai Normal University, Shanghai (2011)Google Scholar