Real-Time Insulator Defect Detection in Overhead Transmission Lines Using YOLOv8

Sabawoon Wali; Jalil Akbarzai; Muhammad Qasim; Hina Zahir; Eamal Shinwari; Habib Ur Rahman Hilal

Authors

Sabawoon Wali Department of Electrical Engineering, Faculty of Electrical and Computer Engineering, University of Engineering and Technology, Peshawar, Pakistan
Jalil Akbarzai Department of Electrical Engineering, Faculty of Electrical and Computer Engineering, University of Engineering and Technology, Peshawar, Pakistan
Muhammad Qasim Department of Electrical Engineering, Faculty of Electrical and Computer Engineering, University of Engineering and Technology, Peshawar, Pakistan
Hina Zahir Department of Electrical Engineering, Faculty of Electrical and Computer Engineering, University of Engineering and Technology, Peshawar, Pakistan
Eamal Shinwari Engineer and Technical Studies Assistant, Operations Directorate, Da Afghanistan Breshna Sherkat (DABS), Afghanistan
Habib Ur Rahman Hilal Department of Computer Science, University of Engineering and Technology, Peshawar, Pakistan

Keywords:

Insulator Defect Detection, YOLOv8m, Deep Learning, Overhead Transmission Lines, Computer Vision, Real-Time Monitoring

Abstract

Insulator defects significantly affect the reliability and safety of overhead transmission lines; hence, their early detection is critical for the stability of the power system. Existing inspection techniques are time-consuming, expensive, and risky, as they involve manual line inspections or helicopter-based systems. This paper presents a real-time insulator defect detection system using the YOLOv8m deep learning model to detect two prominent types of insulator defects: broken insulators and pollution flashovers. A new dataset was created and labeled using Roboflow, and the model was trained and optimized using Google Colab with GPU support. The experimental results indicate that the YOLOv8m model yielded a mean Average Precision (mAP@0.5) of 94.0% and mAP@0.5:0.95 of 68.0%, which is better than the lighter models YOLOv8n and YOLOv8s in terms of detection precision while maintaining real-time performance. The proposed system is a reliable and efficient solution for intelligent inspection and helps in the development of fully automated UAV-based monitoring systems for overhead transmission lines.

References

Md. Ahasan Atick Faisal, Imene Mecheter, Yazan Qiblawey, Javier Hernandez Fernandez, Muhammad E. H. Chowdhury, Serkan Kiranyaz, “Deep Learning in Automated Power Line Inspection: A Review,” arXiv:2502.07826, 2025, [Online]. Available: https://arxiv.org/abs/2502.07826

Yue Liu, Decheng Liu, Xinbo Huang, Chenjing Li, “Insulator defect detection with deep learning: A survey,” IET Gener. Transm. Distrib., 2023, doi: https://doi.org/10.1049/gtd2.12916.

Wenlong Meng, Xuegang Zhang, “Advances in UAV Path Planning: A Comprehensive Review of Methods, Challenges, and Future Directions,” Drones, vol. 9, no. 5, p. 376, 2025, doi: https://doi.org/10.3390/drones9050376.

Qiaodi Wen, Ziqi Luo, “Deep Learning Approaches on Defect Detection in High Resolution Aerial Images of Insulators,” Sensors, vol. 21, no. 4, p. 1033, 2021, doi: https://doi.org/10.3390/s21041033.

Gujing Han, Min He, Feng Zhao, Zhongping Xu, Min Zhang, Liang Qin, “Insulator detection and damage identification based on improved lightweight YOLOv4 network,” Energy Reports, vol. 7, no. 7, pp. 187–197, 2021, doi: https://doi.org/10.1016/j.egyr.2021.10.039.

Jianfeng Zheng, Hang Wu, “Insulator-Defect Detection Algorithm Based on Improved YOLOv7,” Sensors, vol. 22, no. 22, p. 8801, 2022, doi: 10.3390/s22228801.

Jian Ding, Haonan Cao, “High Accuracy Real-Time Insulator String Defect Detection Method Based on Improved YOLOv5,” Front. Energy Res., vol. 10, no. 7, 2022, [Online]. Available: https://www.frontiersin.org/journals/energy-research/articles/10.3389/fenrg.2022.928164/full

Y. Z. Tong Wang, “Insulator Defect Detection Based on ML-YOLOv5 Algorithm,” Sensors, vol. 24, no. 1, p. 204, 2024, doi: https://doi.org/10.3390/s24010204.

Hui He, Xile Huang, Yuxuan Song, Zheng Zhang, Meng Wang, Bo Chen, Guangwei Yan, “An insulator self-blast detection method based on YOLOv4 with aerial images,” Energy Reports, vol. 8, no. 1, pp. 448–454, 2022, doi: https://doi.org/10.1016/j.egyr.2021.11.115.

Y. L. Yifan Qi, “Research on an Insulator Defect Detection Method Based on Improved YOLOv5,” Appl. Sci., vol. 13, no. 9, p. 5741, 2023, doi: https://doi.org/10.3390/app13095741.

Dehong Wei, Bo Hu, “Insulator defect detection based on improved Yolov5s,” Front. Earth Sci., vol. 11, no. 2, 2023, [Online].

Bing Zeng, Zhihao Zhou, Yu Zhou, Dilin He, Zhanpeng Liao, Zihan Jin, Yulu Zhou, Kexin Yi, “An insulator target detection algorithm based on improved YOLOv5,” Sci. Rep., vol. 15, 2025, [Online]. Available: https://www.nature.com/articles/s41598-024-84623-6

“Home - Ultralytics YOLO Docs.” Accessed: Mar. 02, 2026. [Online]. Available: https://docs.ultralytics.com/

Yang Lu, Dahua Li, “A Lightweight Insulator Defect Detection Model Based on Drone Images,” Drones, vol. 8, no. 9, p. 431, 2024, doi: https://doi.org/10.3390/drones8090431.

Michał Tomaszewski, Rafał Gasz, “Detection of Power Line Insulators in Digital Images Based on the Transformed Colour Intensity Profiles,” Sensors, vol. 23, no. 6, p. 3343, 2023, doi: https://doi.org/10.3390/s23063343.

X. Hu and Y. Zhou, “Insulator defect detection in power inspection image using focal loss based on YOLO v4,” p. 30, Dec. 2021, doi: 10.1117/12.2626694.

Tianqi Wang, Yifei Wang, Ping Zong, Shijia Xu, Wei Wei, Hongying Liu, “BMF-SDFormer: A Multi-Scale Feature Fusion Transformer With Spatial-Dual Domain Attention for Insulator Defect Diagnosis,” IET Image Process., 2025, doi: https://doi.org/10.1049/ipr2.70261.

Zhu-ye Xu & Xiao Tang, “Transmission line insulator defect detection algorithm based on MAP-YOLOv8,” Sci. Rep., vol. 15, 2025, [Online]. Available: https://www.nature.com/articles/s41598-025-92445-3

Jun Hu, Wenwei Wan, “Power Insulator Defect Detection Method Based on Enhanced YOLOV7 for Aerial Inspection,” Electronics, vol. 14, no. 3, p. 408, 2025, doi: https://doi.org/10.3390/electronics14030408.

D. Bhardwaj, N. D. Londhe, and R. Raj, “Fault-MTL: A Multi-task Deep Learning Approach for Simultaneous Fault Classification and Localization in Power Systems,” J. Control. Autom. Electr. Syst., vol. 35, no. 5, pp. 884–898, Oct. 2024, doi: 10.1007/s40313-024-01119-4.

Jiangtao Guo, Shu Cao, Tao Wang, Kai Wang, Jingfeng Xiao, Xinxin Meng, “Transformer-based InspecNet for improved UAV surveillance of electrical infrastructure,” Int. J. Appl. Earth Obs. Geoinf., vol. 137, p. 104424, 2025, doi: https://doi.org/10.1016/j.jag.2025.104424.