Tracking Position and Status of Electric Control Switches Based on YOLO Detector

  • Xingang MouEmail author
  • Jian Cui
  • Hujun Yin
  • Xiao Zhou
Conference paper
Part of the Lecture Notes in Computer Science book series (LNCS, volume 11871)


Tracking position and status of switches of electric control cabinets is the key to automatic polling and management systems for intelligent substation. It is a typical task of multi-target image detection and recognition. In this paper, we present an end-to-end switch position detection and state recognition system based on YOLO detector that can detect and recognize multiple targets in a single frame at one time. A four-category network based on YOLOv3-tiny is designed and optimized for real-time detection, then logistic regression is used to predict the probability of status of switches, and then an algorithm based on the prior information of the cabinet is developed to remove duplicate targets. Finally, the detected switches are sorted and numbered to compare with the information in the database. Experiments are reported to verify the proposed system.


Convolutional neural network YOLOv3 State recognition Power systems 



This work was funded by National Natural Science Foundation of China (61701357) and by China Scholarship Council.


  1. 1.
    Zhang, H., Wang, W., Xu, L., Qin, H., Liu, M.: Application of image recognition technology in electrical equipment on-line monitoring. Power Syst. Prot. Control 38(6), 88–91 (2010)Google Scholar
  2. 2.
    Simonyan, K., Zisserman, A.: Very Deep Convolutional Networks for Large-Scale Image Recognition. arXiv preprint arXiv:1409.1556 (2015)
  3. 3.
    Sermanet, P., Eigen, D., Zhang, X.: OverFeat: Integrated Recognition, Localization and Detection using Convolutional Networks. arXiv preprint arXiv:1312.6229 (2014)
  4. 4.
    Uijlings, J.R.R., van de Sande, K.E.A.: Selective search for object recognition. Int. J. Comput. Vis. 104(2), 154–171 (2013)CrossRefGoogle Scholar
  5. 5.
    Girshick, R., Donahue, J., Darrell, T.: Rich feature hierarchies for accurate object detection and semantic segmentation. In: Proceedings of the IEEE International Conference on Computer Vision and Pattern Recognition (CVPR), pp. 580–599 (2014)Google Scholar
  6. 6.
    Girshick, R.: Fast R-CNN. arXiv preprint arXiv:1504.08083 (2015)
  7. 7.
    Ren, S., He, K., Girshick, R., Sun, J.: Faster R-CNN: towards real-time object detection with region proposal networks. In: Proceedings of the 28th International Conference on Neural Information Processing Systems (NIPS), pp. 91–99 (2015)Google Scholar
  8. 8.
    Redmon, J., Divvala, S., Girshick, R.: You only look once: unified, real-time object detection. In: Proceedings of the IEEE International Conference on Computer Vision and Pattern Recognition (CVPR), pp. 779–788 (2015)Google Scholar
  9. 9.
    Liu, W., Anguelov, D., Erhan, D.: SSD: Single Shot MultiBox Detector. arXiv preprint arXiv:1512.02325 (2015)
  10. 10.
    Redmon, J., Farhadi, A.: YOLO9000: better, faster, stronger. In: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR), pp. 7263–7271 (2017)Google Scholar
  11. 11.
    Redmon, J., Farhadi, A.: YOLOv3: An Incremental Improvement. arXiv preprint arXiv:1804.02767 (2018)
  12. 12.
    Bilel, B., Taha, K., Anis K., Adel, A., Kais, O.: Car Detection using Unmanned Aerial Vehicles: Comparison between Faster R-CNN and YOLOv3. arXiv preprint arXiv:1812.10968 (2018)
  13. 13.
    Zhuang, L., Jianguo, L., Zhiqiang S., Gao, H., Shoumeng, Y., Changshui Z.: Learning Efficient Convolutional Networks through Network Slimming. arXiv preprint arXiv:1708.06519 (2017)

Copyright information

© Springer Nature Switzerland AG 2019

Authors and Affiliations

  1. 1.School of Mechanical and Electronic EngineeringWuhan University of TechnologyWuhanChina
  2. 2.School of Electrical and Electronic EngineeringThe University of ManchesterManchesterUK

Personalised recommendations