Abstract
Although great progress has been made in using the method based on deep learning to detect insulators, the detection effect is still unsatisfactory due to the large aspect ratio and inclination of insulators. Therefore, we propose a method capable of detecting large aspect ratio and inclination of insulators, which uses YOLOv3 framework. Firstly, we use manual generation of the anchors to generate angular anchors and fit the label box as much as possible; then we modify the backbone to increase the attention of the detected objects with SE-block; finally, we add angular parameters in rectangular box regression to achieve inclined insulator detection. Compared with the traditional methods, our proposed method significantly improves the performance, and compared with the existing deep learning-based methods, the proposed can achieve real-time detection without degrading performance, and can achieve 82.9% accuracy on our dataset.
This is a preview of subscription content, log in via an institution to check access.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
References
Di, W.: Research on insulator detection and location methods in aerial images. Master’s thesis, North China Electric Power University (2017)
Gao, S.H., Cheng, M.M., Zhao, K., Zhang, X.Y., Yang, M.H., Torr, P.: Res2Net: a new multi-scale backbone architecture. arXiv preprint arXiv:1904.01169 (2019)
Girshick, R.: Fast R-CNN. In: Proceedings of the IEEE International Conference on Computer Vision, pp. 1440–1448 (2015)
He, K., Gkioxari, G., Dollár, P., Girshick, R.: Mask R-CNN. In: Proceedings of the IEEE International Conference on Computer Vision, pp. 2961–2969 (2017)
He, K., Zhang, X., Ren, S., Sun, J.: Deep residual learning for image recognition. In: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, pp. 770–778 (2016)
Hu, J., Shen, L., Sun, G.: Squeeze-and-excitation networks. In: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, pp. 7132–7141 (2018)
Huang, G., Liu, Z., Van Der Maaten, L., Weinberger, K.Q.: Densely connected convolutional networks. In: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, pp. 4700–4708 (2017)
Jaderberg, M., Simonyan, K., Zisserman, A., et al.: Spatial transformer networks. In: Advances in Neural Information Processing Systems, pp. 2017–2025 (2015)
Jiang, Y., et al.: R2CNN: rotational region CNN for orientation robust scene text detection. arXiv preprint arXiv:1706.09579 (2017)
Ling, Z., Qiu, R.C., Jin, Z., Zhang, Y., Lei, C.: An accurate and real-time self-blast glass insulator location method based on faster R-CNN and U-net with aerial images. arXiv preprint arXiv:1801.05143 (2018)
Ma, J., et al.: Arbitrary-oriented scene text detection via rotation proposals. IEEE Trans. Multimedia 20(11), 3111–3122 (2018)
Neubeck, A., Gool, L.V.: Efficient non-maximum suppression. In: International Conference on Pattern Recognition (2006)
Redmon, J., Farhadi, A.: YOLOv3: an incremental improvement. arXiv preprint arXiv:1804.02767 (2018)
Ren, S., He, K., Girshick, R., Sun, J.: Faster R-CNN: towards real-time object detection with region proposal networks. In: Advances in Neural Information Processing Systems, pp. 91–99 (2015)
Simonyan, K., Zisserman, A.: Very deep convolutional networks for large-scale image recognition. arXiv preprint arXiv:1409.1556 (2014)
Singh, B., Davis, L.S.: An analysis of scale invariance in object detection snip. In: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, pp. 3578–3587 (2018)
Tao, X., Zhang, D., Wang, Z., Liu, X., Xu, D.: Detection of power line insulator defects using aerial images analyzed with convolutional neural networks. IEEE Trans. Syst. Man Cybern.: Syst., 1–13 (2018)
Wang, F., et al.: Residual attention network for image classification. In: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, pp. 3156–3164 (2017)
Wu, Y.: Study on insulator identification and state detection method based on aerial image. Ph.D. thesis, North China Electric Power University (2016)
Xia, F., Tie, H.F., Shi, E.W., Zhang, Q., Wang, L.: Insulator recognition based on mathematical morphology and Bayesian segmentation. In: International Congress on Image & Signal Processing (2018)
Yang, X., et al.: R2CNN++: multi-dimensional attention based rotation invariant detector with robust anchor strategy. arXiv preprint arXiv:1811.07126 (2018)
Yang, X., Sun, H., Sun, X., Yan, M., Guo, Z., Fu, K.: Position detection and direction prediction for arbitrary-oriented ships via multitask rotation region convolutional neural network. IEEE Access 6, 50839–50849 (2018)
Yue, L., Yong, J., Liang, L., Zhao, J., Li, Z.: The method of insulator recognition based on deep learning. In: International Conference on Applied Robotics for the Power Industry (2016)
Yuwei, H.: Extraction and recognition of aerial video insulator parts. Ph.D. thesis, Dalian Maritime University (2013)
Acknowledgments
We would like to thank the youth program of National Natural Science Foundation of China (Grant No. 61703060), science and technology program of Sichuan Province (Grant No. 2019yj0165) and general program of National Natural Science Foundation of China (Grant No. 61973055).
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2020 Springer Nature Switzerland AG
About this paper
Cite this paper
Ye, R., Wang, Q., Yan, B., Li, B. (2020). Position and Orientation Detection of Insulators in Arbitrary Direction Based on YOLOv3. In: Peng, Y., et al. Pattern Recognition and Computer Vision. PRCV 2020. Lecture Notes in Computer Science(), vol 12305. Springer, Cham. https://doi.org/10.1007/978-3-030-60633-6_46
Download citation
DOI: https://doi.org/10.1007/978-3-030-60633-6_46
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-030-60632-9
Online ISBN: 978-3-030-60633-6
eBook Packages: Computer ScienceComputer Science (R0)