Method for determining the inclination of a communication line pole based on UAV images

Purpose of research. Overhead communication lines (OCL) are an important element of the communication infrastructure, but their technical condition requires regular monitoring and inspection. Traditional inspection methods, including visual inspection by specialists, do not always allow for the efficient collection and recording of all necessary data. In order to improve the quality of OCL inspection, a method was developed for determining the tilt of a communication line pole based on images from an unmanned aerial vehicle (UAV).

Methods. A combination of mathematical transformations and machine learning methods was used to solve the problem. Data processing included the use of camera parameters, object coordinates in the image, flight altitude, and UAV coordinates. Based on these data, an algorithm was developed for detecting key support points and calculating the tilt angle of the poles.

Results. As a result of the experiments conducted based on the data obtained from the UAV, the accuracy of detecting key support points according to the mAP50 metric was 0.71. Within the correctly predicted support, the accuracy of detecting its top and base was 0.88 according to the F1-score metric. To determine the tilt of the VLS pillars, a formula was derived that made it possible to calculate the maximum tilt of the pillar is 24.5°, and the minimum is 0.6°. The average tilt angle of the pillars for the entire set of images is approximately 6.1°.

Conclusion. The developed method allows automating the technical inspection of VLS, ensuring high accuracy in determining their key parameters. The use of UAVs and machine learning reduces time and cost, and improves the quality of data collection and analysis. The use of UAVs in combination with machine learning methods can significantly reduce time and cost, improve the quality of data collection and analysis, and reduce the risk of human error.

1. Astapova M.A., Uzdaev M.Yu. Detection of defects in faulty elements of power lines using YOLO neural networks. Modelirovanie, optimizatsiya i informatsionnye tekhnologii = Modeling, Optimization and Information Technologies. 2021; 9(4): 35. (In Russ.).

2. Pankov D.N., Bugorskiy I.A. Application of artificial intelligence for defect detection in overhead power lines. Perspektivy razvitiya tekhnologij obrabotki i oborudovaniya v mashinostroenii = Perspectives of development of technologies for processing and equipment in mechanical engineering. 2023; 282–285. (In Russ.).

3. Fang Z., Savkin A.V. Strategies for Optimized UAV Surveillance in Various Tasks and Scenarios: A Review. Drones. 2024; 8(5): 193.

4. Epifani L., Caruso A. A Survey on Deep Learning in UAV Imagery for Precision Agriculture and Wild Flora Monitoring: Datasets, Models and Challenges. Smart Agricultural Technology. 2024; 100625.

5. Yılmaz Y., Gürtürk M., Süleymanoğlu B., Soycan A., Soycan M. Overview of Large Scale Map Production with UAV-Based Photogrammetric Technique: A Case Study in Izmir-Cesme Territory of Turkey. Journal of Geography and Cartography. 2023; 6(2): 2164.

6. Tsellou A., Livanos G., Ramnalis D., Polychronos V., Plokamakis G., Zervakis M., Moirogiorgou K. A UAV Intelligent System for Greek Power Lines Monitoring. Sensors. 2023; 23(20): 8441.

7. Bellou E., Pisica I., Banitsas K. Aerial Inspection of High-Voltage Power Lines Using YOLOv8 Real-Time Object Detector. Energies. 2024; 17(11): 2535.

8. Remondino F., Barazzetti L., Nex F., Scaioni M., Sarazzi D. UAV Photogrammetry for Mapping and 3D Modeling: Current Status and Future Perspectives. Proceedings of the International Conference on Unmanned Aerial Vehicle in Geomatics (UAV-g). 2011; 25–31.

9. Bhowmick S., Nagarajaiah S., Veeraraghavan A. Vision and Deep Learning-Based Algorithms to Detect and Quantify Cracks on Concrete Surfaces from UAV Videos. Sensors. 2020; 20(21): 6299.

10. Chen L., Chang J., Xu J., Yang Z. Automatic Measurement of Inclination Angle of Utility Poles Using 2D Image and 3D Point Cloud. Applied Sciences. 2023; 13(3): 1688.

11. Alam M.M., Zhu Z., Eren Tokgoz B., Zhang J., Hwang S. Automatic Assessment and Prediction of the Resilience of Utility Poles Using Unmanned Aerial Vehicles and Computer Vision Techniques. International Journal of Disaster Risk Science. 2020; 11: 119–132.

12. Chen F., Li Y., Shuang F., Huang M. PTTE: Power Tower Tilt Estimation Algorithm Based on LiDAR Point Cloud. 2023 International Conference on Advanced Robotics and Mechatronics (ICARM). IEEE. 2023; 309–315.

13. Limanovskaya O.V., Titov E.A., Volkova D.I., Lemeh A.V. Algorithm for determining the inclination of power line supports using deep learning methods on video data. Vestnik Ivanovskogo gosudarstvennogo energeticheskogo universiteta = Bulletin of Ivanovo State Power University. 2020; (2): 72–80. (In Russ.).

14. Zheng Y., Wu Y., Gao J., Cui S. NB-IoT Based Method for Monitoring the Tilt Status of Transmission Towers. Journal of Physics: Conference Series. IOP Publishing. 2021; 2108(1): 012033.

15. Lu Z., Gong H., Jin Q., Hu Q., Wang S. A Transmission Tower Tilt State Assessment Approach Based on Dense Point Cloud from UAV-Based LiDAR. Remote Sensing. 2022; 14(2): 408.

16. Zhu Z., Zhang J., Alam M.M., Tokgoz B.E., Hwang S. Automatic Utility Pole Inclination Angle Measurement Using Unmanned Aerial Vehicle and Deep Learning. Proceedings of the Institute of Industrial and Systems Engineers Annual Conference. 2019.

17. Sun L., Ma C., Hu H., Ni T., Zhang M., Zhang Z. Power Tower Tilt Monitoring Method Based on Beidou and Improved YOLOv3. AIoTC. 2022; 58–66.

18. Yue K. Multi-Sensor Data Fusion for Autonomous Flight of Unmanned Aerial Vehicles in Complex Flight Environments. Drone Systems and Applications. 2024; 12: 1–12.

19. Sun H., Shen Q., Ke H., Duan Z., Tang X. Power Transmission Lines Foreign Object Intrusion Detection Method for Drone Aerial Images Based on Improved YOLOv8 Network. Drones. 2024; 8(8): 346.

20. Chi X., Sun Y., Zhao Y., Lu D., Gao Y., Zhang Y. An Improved YOLOv8 Network for Detecting Electric Pylons Based on Optical Satellite Image. Sensors. 2024; 24(12): 4012.

21. Maji D., Nagori S., Mathew M., Poddar D. YOLO-Pose: Enhancing YOLO for Multi-Person Pose Estimation Using Object Keypoint Similarity Loss. In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition. 2022; 2637–2646.

22. Farnebäck G. Two-Frame Motion Estimation Based on Polynomial Expansion. In: Image Analysis: 13th Scandinavian Conference, SCIA 2003, Halmstad, Sweden, June 29– July 2, 2003 Proceedings 13. Springer Berlin Heidelberg. 2003; 363–370.