10.57647/j.mjee.2025.17412

An Improved Deep Mask Region-based Convolutional Neural Network for the Detection of Power Transmission Insulator Defects

PDF
  • Haider Abdulzahra Saad Alsaide1,
  • Mohammadreza Soltanaghaei*1,
  • Wael Hussein Zayer Al-Lami2,
  • Razieh Asgarnezhad1,
  1. Department of Computer Engineering, Institute of Artificial Intelligence and Social and Advanced Technologies, Isf.C., Islamic Azad University, Isfahan, Iran
  2. Electronic Department, Amara Technical Institute, Southern Technical University, Missan, Iraq

Received: 2025-03-15

Revised: 2025-06-14

Accepted: 2025-07-07

Published in Issue 2025-09-01

Copyright (c) 2025 Haider Abdulzahra Saad Alsaide, Mohammadreza Soltanaghaei, Wael Hussein Zayer Al-Lami, Razieh Asgarnezhad (Author)

Creative Commons License

This work is licensed under a Creative Commons Attribution 4.0 International License.

How to Cite

Alsaide, H. A. S., Soltanaghaei, M., Al-Lami, W. H. Z., & Asgarnezhad, R. (2025). An Improved Deep Mask Region-based Convolutional Neural Network for the Detection of Power Transmission Insulator Defects. Majlesi Journal of Electrical Engineering. https://doi.org/10.57647/j.mjee.2025.17412
Crossref
Scopus
Google Scholar
Europe PMC

PDF views: 38

Abstract

In the power industry, accurate and real-time defect detection in power transmission lines is crucial to reducing accidents and maintenance costs. Although various image processing and AI-based methods have been proposed using unmanned arial vehicles (UAV) images, their efficiency significantly drops under complex environmental conditions, such as high color or texture diversity in backgrounds. To address this challenge, this paper presents an improved version of the mask region-based convolutional neural network (R-CNN) network by redefining the architecture of the head and classification sections to enhance detection accuracy in complex scenarios. The proposed model introduces a novel rhombus-structured fully connected layer in the classification branch for better feature encoding and decoding, alongside a drop-out layer in the bounding box detection branch to prevent overfitting. Additionally, ResNet-101 is employed as the backbone, and transfer learning is used to optimize training and reduce computational complexity. Experimental evaluations demonstrate that the improved Mask R-CNN achieves a mean average precision (mAP) of 92.07% and an accuracy of 92.50% in detecting power transmission line defects, outperforming existing methods. Furthermore, the proposed approach helps lower the cost and time required for fault detection and maintenance, making it a practical solution for real-world power transmission system inspections.

Keywords

  • Power transmission line,
  • Defect detection,
  • Deep learning,
  • Feature extraction,
  • Mask region-based convolutional neural network,
  • Transfer learning
PDF

References

  1. Ayala García, A., Galván Bobadilla, I., Arroyo Figueroa, G., Pérez Ramírez, M., & Muñoz Román, J. (2016). Virtual reality training system for maintenance and operation of high-voltage overhead power lines. Virtual Reality, 20, 27-40.
  2. Liu, J., Hu, M., Dong, J., & Lu, X. (2023). Summary of insulator defect detection based on deep learning. Electric Power Systems Research, 224, 109688.
  3. Mirallès, F., Pouliot, N., & Montambault, S. (2014, October). State-of-the-art review of computer vision for the management of power transmission lines. In Proceedings of the 2014 3rd international conference on applied robotics for the power industry (pp. 1-6). IEEE.
  4. Zhou, Z., Zhang, C., Xu, C., Xiong, F., Zhang, Y., & Umer, T. (2018). Energy-efficient industrial Internet of UAVs for power line inspection in smart grid. IEEE Transactions on Industrial Informatics, 14(6), 2705-2714.
  5. Ni, L., Ma, Y., Lin, Q., Yang, J., & Jin, L. (2021, October). Research on Insulator Defect Detection Method Based on Image Processing and Watershed Algorithm. In 2021 International Conference on Advanced Electrical Equipment and Reliable Operation (AEERO) (pp. 1-6). IEEE.
  6. Xia, H., Yang, B., Li, Y., & Wang, B. (2022). An improved CenterNet model for insulator defect detection using aerial imagery. Sensors, 22(8), 2850.
  7. He, K., Gkioxari, G., Dollár, P., & Girshick, R. (2017). Mask r-cnn. In Proceedings of the IEEE international conference on computer vision (pp. 2961-2969).
  8. Girshick, R. (2015). Fast r-cnn. In Proceedings of the IEEE international conference on computer vision (pp. 1440-1448)
  9. Jabid, T., & Ahsan, T. (2018). Insulator detection and defect classification using rotation invariant local directional pattern. International Journal of Advanced Computer Science and Applications, 9(2).
  10. Iruansi, U., Tapamo, J. R., & Davidson, I. E. (2019). Classification of power-line insulator condition using local binary patterns with support vector machines. IAENG International Journal of Computer Science, 46(2), 300-310.
  11. Siddiqui, Z. A., Park, U., Lee, S. W., Jung, N. J., Choi, M., Lim, C., & Seo, J. H. (2018). Robust powerline equipment inspection system based on a convolutional neural network. Sensors, 18(11), 3837.
  12. Han, J., Yang, Z., Xu, H., Hu, G., Zhang, C., Li, H., ... & Zeng, H. (2020). Search like an eagle: A cascaded model for insulator missing faults detection in aerial images. Energies, 13(3), 713.
  13. Panigrahy, S., Karmakar, S., & Sahoo, R. (2021, July). Condition Assessment of High Voltage Insulator using Convolutional Neural Network. In 2021 IEEE International Conference on Electronics, Computing and Communication Technologies (CONECCT) (pp. 1-6). IEEE.
  14. Chen, W., Li, Y., & Zhao, Z. (2021). InsulatorGAN: A transmission line insulator detection model using multi-granularity conditional generative adversarial nets for UAV inspection. Remote Sensing, 13(19), 3971.
  15. Liu, C., Wu, Y., Liu, J., Sun, Z., & Xu, H. (2021). Insulator faults detection in aerial images from high-voltage transmission lines based on deep learning model. Applied Sciences, 11(10), 4647.
  16. Liquan, Z., Mengjun, Z., Ying, C., & Yanfei, J. (2022). Fast detection of defective insulator based on improved YOLOv5s. Computational Intelligence and Neuroscience, 2022.
  17. Dong, C., Zhang, K., Xie, Z., & Shi, C. (2023). An improved cascade RCNN detection method for key components and defects of transmission lines. IET Generation, Transmission & Distribution, 17(19), 4277-4292.
  18. Sui, T., & Wang, J. (2023). PDDD-Net: Defect Detection Network based on Parallel Attention Mechanism and Dual-Channel Spatial Pyramid Pooling. IEEE Access.
  19. Elngar, A. A., Arafa, M., Fathy, A., Moustafa, B., Mahmoud, O., Shaban, M., & Fawzy, N. (2021). Image classification based on CNN: a survey. Journal of Cybersecurity and Information Management, 6(1), 18-50.
  20. Bharati, P., & Pramanik, A. (2020). Deep learning techniques—R-CNN to mask R-CNN: a survey. Computational Intelligence in Pattern Recognition: Proceedings of CIPR 2019, 657-668.
  21. Girshick, R. (2015). Fast r-cnn. In Proceedings of the IEEE international conference on computer vision (pp. 1440-1448).
  22. He, K., Gkioxari, G., Dollár, P., & Girshick, R. (2017). Mask r-cnn. In Proceedings of the IEEE international conference on computer vision (pp. 2961-2969).
  23. Wu, M., Yue, H., Wang, J., Huang, Y., Liu, M., Jiang, Y., Zeng, C. (2020). Object detection based on RGC mask R‐CNN. IET Image Processing, 14(8), 1502-1508.
  24. Tian, Y., Yang, G., Wang, Z., Li, E., & Liang, Z. (2020). Instance segmentation of apple flowers using the improved mask R–CNN model. Biosystems engineering, 193, 264-278.
  25. Zhang, Y., Liu, Y. L., Nie, K., Zhou, J., Chen, Z., Chen, J. H., ... & Su, M. Y. (2023). Deep learning-based automatic diagnosis of breast cancer on MRI using mask R-CNN for detection followed by ResNet50 for classification. Academic radiology, 30, S161-S171.
  26. Zhang, Y., Liu, Y. L., Nie, K., Zhou, J., Chen, Z., Chen, J. H., ... & Su, M. Y. (2023). Deep learning-based automatic diagnosis of breast cancer on MRI using mask R-CNN for detection followed by ResNet50 for classification. Academic radiology, 30, S161-S171.
  27. Wang, Y., Wang, X., Hao, R., Lu, B., & Huang, B. (2024). Metal Surface Defect Detection Method Based on Improved Cascade R-CNN. Journal of Computing and Information Science in Engineering, 24(4), 041002.
  28. Dash, S. P., Ramadevi, J., Amat, R., Sethy, P. K., Behera, S. K., & Mallick, S. (2023, September). Wafer Defect Identification with Optimal Hyper-Parameter Tuning of Support Vector Machine using the Deep Feature of ResNet 101. In IOP Conference Series: Materials Science and Engineering (Vol. 1291, No. 1, p. 012048). IOP Publishing.
  29. Yu, H., Miao, X., & Wang, H. (2022). Bearing fault reconstruction diagnosis method based on ResNet-152 with multi-scale stacked receptive field. Sensors, 22(5), 1705.
  30. Munir, N., Kim, H. J., Song, S. J., & Kang, S. S. (2018). Investigation of deep neural network with drop out for ultrasonic flaw classification in weldments. Journal of Mechanical Science and Technology, 32, 3073-3080.
  31. Waleed Abdulla. Mask R-CNN for object detection and instance segmentation on Keras and TensorFlow. GitHub repository. Available online: https://github.com/matterport/Mask_RCNN (accessed on 10 January 2024).
  32. IEEE Dataport. Insulator Defect Detection Dataset - Competition. Available online: https://ieee-dataport.org/competitions/insulator-defect-detection (accessed on 14 January 2024).
  33. Qiu, Z., Zhu, X., Liao, C., Shi, D., & Qu, W. (2022). Detection of transmission line insulator defects based on an improved lightweight YOLOv4 model. Applied Sciences, 12(3), 1207.
  34. Chen, Y., Liu, H., Chen, J., Hu, J., & Zheng, E. (2023). Insu-YOLO: an insulator defect detection algorithm based on multiscale feature fusion. Electronics, 12(15), 3210.
  35. Zhang, T., Zhang, Y., Xin, M., Liao, J., & Xie, Q. (2023). A light-weight network for small insulator and defect detection using UAV imaging based on improved YOLOv5. Sensors, 23(11), 5249.
  36. Chen, B., Zhang, W., Wu, W., Li, Y., Chen, Z., & Li, C. (2024). ID-YOLOv7: an efficient method for insulator defect detection in power distribution network. Frontiers in Neurorobotics, 17, 1331427.
  37. Chen, Y., Deng, C., Sun, Q., Wu, Z., Zou, L., Zhang, G., & Li, W. (2024). Lightweight Detection Methods for Insulator Self-Explosion Defects. Sensors, 24(1), 290.
  38. Wang, C. Y., Bochkovskiy, A., & Liao, H. Y. M. (2023). YOLOv7: Trainable bag-of-freebies sets new state-of-the-art for real-time object detectors. In Proceedings of the IEEE/CVF conference on computer vision and pattern recognition (pp. 7464-7475).
  39. Moradi, E. (2025). Accuracy Enhancement of Fault Diagnosis for Power Transformers with a Hybrid Approach Integrating Robust and Tree-Based Algorithms. Majlesi Journal of Electrical Engineering, 19(1), 1–9.
  40. Ghahraei, O., Kheshtzarrin, M., & Amiri, M. (2023). Development of a Button Mushroom Harvesting Robot Using Expert System and Image Processing in Shelf Cultivation Method. Majlesi Journal of Electrical Engineering, 17(2), 45–52.
  41. Rezaee, K., Ahmadi, M. K. N., & Anari, M. S. (2022). Interactive Medical Image Segmentation Using Active Contour with Improved F Energy in Level-Set Tuning. Majlesi Journal of Electrical Engineering, 16(4), 210–218.
  42. Kalluri, R. D., & Selvaraj, P. (2024). A Review on Application of Various Deep Learning Techniques and Filtering Approach in Plant Phenotyping. Majlesi Journal of Electrical Engineering, 18(2), 103–110.
  43. Jiang, Y., Cheng, T., Dong, J., Liang, J., Zhang, Y., Lin, X., & Yao, H. (2022). Dermoscopic image segmentation based on Pyramid Residual Attention Module. PLoS ONE, 17(9), e0267380. https://doi.org/10.1371/journal.pone.0267380
  44. Sun, Y., Malihi, S., Li, H., & Maboudi, M. (2022). DeepWindows: Windows instance segmentation through an improved Mask R-CNN using spatial attention and relation modules. ISPRS International Journal of Geo-Information, 11(3), 162. https://doi.org/10.3390/ijgi11030162.
  45. He, M., He, K., Huang, Q., Xiao, H., Zhang, H., Li, G., & Chen, A. (2025). Lightweight Mask R-CNN for instance segmentation and particle physical property analysis in multiphase flow. Powder Technology, 449, 120366. https://doi.org/10.1016/j.powtec.2024.120366.