Potential analyses of LiDAR-based automatic powerline detection algorithms


  • Mertcan Nazar
  • Umut Gunes Sefercik
  • Ilyas Aydın


Powerline detection, LiDAR, Point cloud, Aerial systems, Disaster monitoring


Powerlines are utilized for distributing electricity for household use, industry, healthcare purposes and etc. The detection of powerlines is an important topic in object detection studies. In addition, overgrown trees in dense forest regions may pose a risk for powerlines positioned in these regions and even may cause forest fires if unattended. So, it is also important in the scope of disaster monitoring research and studies. Light detection and ranging (LiDAR) systems are utilized for the detection of powerlines in urban regions and forests with the ability to obtain high-resolution point cloud data. Also, with the operation principle based on active remote sensing aerial LiDAR systems with multi-return capability can be used to obtain information about forest understory and is more effective compared to optical systems in this context. In this study, aerial LiDAR point cloud data of an urban region was utilized for the automatic detection of powerlines. For the automatic detection of powerlines, Robust Railroad Infrastructure Detection Framework which was developed by Eötvös Loránd University (ELTE) Geoinformatics Laboratory was utilized and five algorithms including Above, AngleAbove, AngleGroundAbove, VoronoiAbove, and VoronoiGroundAbove are applied separately on LiDAR point cloud data. When results are analyzed visually AngleAbove gave the best results in powerline detection.  


Mitchell, J. W. (2009, January). Power lines and catastrophic wildland fire in Southern California. In Proceedings of the 11th International Conference on Fire and Materials (pp. 225-238).

Jwa, Y., Sohn, G., & Kim, H. B. (2009). Automatic 3d powerline reconstruction using airborne lidar data. The International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences, 38(Part 3), W8.

Sefercik, U. G., Glennie, C., Singhania, A., & Hauser, D. (2015). Area-based quality control of airborne laser scanning 3D models for different land classes using terrestrial laser scanning: sample survey in Houston, USA. International Journal of Remote Sensing, 36(23), 5916-5934.

Sefercik, U. G., Buyuksalih, G., Jacobsen, K., & Alkan, M. (2017). Point-based and model-based geolocation analysis of airborne laser scanning data. Optical Engineering, 56(1), 013101, 1-10

Koetz, B., Morsdorf, F., Van der Linden, S., Curt, T., & Allgöwer, B. (2008). Multi-source land cover classification for forest fire management based on imaging spectrometry and LiDAR data. Forest Ecology and Management, 256(3), 263-271.

Hervo, M., Quennehen, B., Kristiansen, N. I., Boulon, J., Stohl, A., Fréville, P., ... & Sellegri, K. (2012). Physical and optical properties of 2010 Eyjafjallajökull volcanic eruption aerosol: ground-based, Lidar and airborne measurements in France. Atmospheric Chemistry and Physics, 12(4), 1721-1736.

Moya, L., Yamazaki, F., Liu, W., & Yamada, M. (2018). Detection of collapsed buildings from lidar data due to the 2016 Kumamoto earthquake in Japan. Natural Hazards and Earth System Sciences, 18(1), 65-78.

Zhu, L., & Hyyppä, J. (2014). Fully-automated power line extraction from airborne laser scanning point clouds in forest areas. Remote Sensing, 6(11), 11267-11282.

Cserép, M., Hudoba, P., & Vincellér, Z. (2018). Robust Railroad Cable Detection in Rural Areas from MLS Point Clouds. In Free and Open Source Software for Geospatial (FOSS4G) Conference Proceedings, 18(1), 1-8

Stephens, R. S. (1991). Probabilistic approach to the Hough transform. Image and vision computing, 9(1), 66-71.