Optimized YOLOv7-tiny model for smoke detection in power transmission lines

Chen Chen; Guowu Yuan; Hao Zhou; Yutang Ma; Yi Ma; Chen Chen; Guowu Yuan; Hao Zhou; Yutang Ma; Yi Ma

doi:10.3934/mbe.2023853

Mathematical Biosciences and Engineering

2023, Volume 20, Issue 11: 19300-19319. doi: 10.3934/mbe.2023853

Previous Article Next Article

Research article Special Issues

Optimized YOLOv7-tiny model for smoke detection in power transmission lines

1.
School of Information Science and Engineering, Yunnan University, Kunming 650504, Yunnan, China
2.
Electric Power Research Institute, Yunnan Power Grid Co., Ltd, Kunming 650217, Yunnan, China
3.
Yunnan Key Laboratory of Intelligent Systems and Computing, Kunming 650504, Yunnan, China

Academic Editor: Shangce Gao

Received: 08 August 2023 Revised: 09 September 2023 Accepted: 17 September 2023 Published: 17 October 2023

Fire incidents near power transmission lines pose significant safety hazards to the regular operation of the power system. Therefore, achieving fast and accurate smoke detection around power transmission lines is crucial. Due to the complexity and variability of smoke scenarios, existing smoke detection models suffer from low detection accuracy and slow detection speed. This paper proposes an improved model for smoke detection in high-voltage power transmission lines based on the improved YOLOv7-tiny. First, we construct a dataset for smoke detection in high-voltage power transmission lines. Due to the limited number of real samples, we employ a particle system to randomly generate smoke and composite it into randomly selected real scenes, effectively expanding the dataset with high quality. Next, we introduce multiple parameter-free attention modules into the YOLOv7-tiny model and replace regular convolutions in the Neck of the model with Spd-Conv (Space-to-depth Conv) to improve detection accuracy and speed. Finally, we utilize the synthesized smoke dataset as the source domain for model transfer learning. We pre-train the improved model and fine-tune it on a dataset consisting of real scenarios. Experimental results demonstrate that the proposed improved YOLOv7-tiny model achieves a 2.61% increase in mean Average Precision (mAP) for smoke detection on power transmission lines compared to the original model. The precision is improved by 2.26%, and the recall is improved by 7.25%. Compared to other object detection models, the smoke detection proposed in this paper achieves high detection accuracy and speed. Our model also improved detection accuracy on the already publicly available wildfire smoke dataset Figlib (Fire Ignition Library).
- transmission line,
- smoke detection,
- attention mechanism,
- deep learning
Citation: Chen Chen, Guowu Yuan, Hao Zhou, Yutang Ma, Yi Ma. Optimized YOLOv7-tiny model for smoke detection in power transmission lines[J]. Mathematical Biosciences and Engineering, 2023, 20(11): 19300-19319. doi: 10.3934/mbe.2023853

Related Papers:

Abstract

Fire incidents near power transmission lines pose significant safety hazards to the regular operation of the power system. Therefore, achieving fast and accurate smoke detection around power transmission lines is crucial. Due to the complexity and variability of smoke scenarios, existing smoke detection models suffer from low detection accuracy and slow detection speed. This paper proposes an improved model for smoke detection in high-voltage power transmission lines based on the improved YOLOv7-tiny. First, we construct a dataset for smoke detection in high-voltage power transmission lines. Due to the limited number of real samples, we employ a particle system to randomly generate smoke and composite it into randomly selected real scenes, effectively expanding the dataset with high quality. Next, we introduce multiple parameter-free attention modules into the YOLOv7-tiny model and replace regular convolutions in the Neck of the model with Spd-Conv (Space-to-depth Conv) to improve detection accuracy and speed. Finally, we utilize the synthesized smoke dataset as the source domain for model transfer learning. We pre-train the improved model and fine-tune it on a dataset consisting of real scenarios. Experimental results demonstrate that the proposed improved YOLOv7-tiny model achieves a 2.61% increase in mean Average Precision (mAP) for smoke detection on power transmission lines compared to the original model. The precision is improved by 2.26%, and the recall is improved by 7.25%. Compared to other object detection models, the smoke detection proposed in this paper achieves high detection accuracy and speed. Our model also improved detection accuracy on the already publicly available wildfire smoke dataset Figlib (Fire Ignition Library).

References

[1]	Z. B. Zhao, Z. G. Jiang, Y. X. Li, Y. C. Qi, Y. J. Zhai, W. Q. Zhao, et al., Overview of visual defect detection of transmission line components, J. Image Graphics, 26 (2021), 2545–2560. https://doi.org/10.11834/jig.200689 doi: 10.11834/jig.200689
[2]	Y. Sui, P. F. Ning, P. J. Niu, C. Y. Wang, D. Zhao, W. L. Zhang, et al, Review on mounted UAV for transmission line inspection, Power Syst. Technol., 45 (2021), 3636–3648. http://doi.org/10.13335/j.1000-3673.pst.2020.1178 doi: 10.13335/j.1000-3673.pst.2020.1178
[3]	Z. Y. Liu, X. R. Miu, J. Chen, H. Jiang, Review of visible image intelligent processing for transmission line inspection, Power Syst. Technol., 44 (2020), 1057–1069. http://doi.org/10.13335/j.1000-3673.pst.2019.0349 doi: 10.13335/j.1000-3673.pst.2019.0349
[4]	S. Khan, K. Muhammad, S. Mumtaz, S. W. Baik, V. H. C. Albuquerque, Energy-efficient deep CNN for smoke detection in foggy IoT environment, IEEE Internet Things J., 6 (2019), 9237–9245. http://doi.org/10.1109/JIOT.2019.2896120 doi: 10.1109/JIOT.2019.2896120
[5]	H. Yin, Y. R. Wei, An improved algorithm based on convolutional neural network for smoke detection, in 2019 IEEE International Conferences on Ubiquitous Computing & Communications (IUCC) and Data Science and Computational Intelligence (DSCI) and Smart Computing, Networking and Services (SmartCNS), IEEE, (2019), 207–211. http://doi.org/10.1109/IUCC/DSCI/SmartCNS.2019.00063
[6]	C. H. Li, B. Yang, H. Ding, H. L. Shi, X. P. Jiang, J. Sun, Real-time video-based smoke detection with high accuracy and efficiency, Fire Saf. J., 117 (2020), 103184. http://doi.org/10.1016/j.firesaf.2020.103184 doi: 10.1016/j.firesaf.2020.103184
[7]	M. H. Jiang, Y. X. Zhao, F. Yu, C. L. Zhou, T. Peng, A self-attention network for smoke detection, Fire Saf. J., 129 (2022), 103547. http://doi.org/10.1016/j.firesaf.2022.103547 doi: 10.1016/j.firesaf.2022.103547
[8]	Z. Q. Li, A. Khananian, R. H. Fraser, J. Cihlar, Automatic detection of fire smoke using artificial neural networks and threshold approaches applied to AVHRR imagery, IEEE Trans. Geosci. Remote Sens., 39 (2001), 1859–1870. http://doi.org/10.1109/36.951076 doi: 10.1109/36.951076
[9]	K. Muhammad, J. Ahmad, I. Mehmood, S. Rho, S. W. Baik, Convolutional neural networks based fire detection in surveillance videos, IEEE Access, 6 (2018), 18174–18183. http://10.1109/ACCESS.2018.2812835 doi: 10.1109/ACCESS.2018.2812835
[10]	W. B. Cai, C. Y. Wang, H. Huang, T. Z. Wang, A real-time smoke detection model based on YOLO-smoke algorithm, in 2020 Cross Strait Radio Science & Wireless Technology Conference (CSRSWTC), IEEE, (2020), 1–3. http://10.1109/CSRSWTC50769.2020.9372453
[11]	F. R. Zhou, G. Wen, Y. Ma, Y. F. Wang, Y. T. Ma, G. F. Wang, et al., Multilevel feature cooperative alignment and fusion for unsupervised domain adaptation smoke detection, Front. Phys., 11 (2023), 81. https://doi.org/10.3389/fphy.2023.1136021 doi: 10.3389/fphy.2023.1136021
[12]	S. G. Zhang, F. Zhang, Y. Ding, Y. Li, Swin-YOLOv5: Research and application of fire and smoke detection algorithm based on YOLOv5, Comput. Intell. Neurosci., 2022 (2022). https://doi.org/10.1155/2022/6081680 doi: 10.1155/2022/6081680
[13]	C. Y. Wang, A. Bochkovskiy, H. Y. M. Liao, 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, (2023), 7464–7475. https://doi.org/10.48550/arXiv.2207.02696
[14]	Y. C. Zhou, L. H. Fang, X. Y. Zheng, X. L. Chen, Virtual battlefield smoke effect simulation based on particle system, Comput. Simul., 32 (2015), 417–420. https://doi.org/10.3969/j.issn.1006-9348.2015.07.093 doi: 10.3969/j.issn.1006-9348.2015.07.093
[15]	A. Bochkovskiy, C. Y. Wang, H. Y. M. Liao, YOLOv4: Optimal speed and accuracy of object detection, preprint, arXiv: 2004.10934. https://doi.org/10.48550/arXiv.2004.10934
[16]	G. Jocher, A. Stoken, J. Borovec, L. Changyu, A. Hogan, L. Diaconu, et al., ultralytics/yolov5: v3. 0, Zenodo, 2020. Available from: https://ui.adsabs.harvard.edu/abs/2020zndo...3983579J/abstract.
[17]	Y. Liu, X. Wang, Sar ship detection based on improved YOLOv7-tiny, in 2022 IEEE 8th International Conference on Computer and Communications (ICCC), IEEE, (2022), 2166–2170. https://doi.org/10.1109/ICCC56324.2022.10065775
[18]	T. Y. Lin, P. Dollár, R. Girshick, K. He, B. Hariharan, S. Belongie, Feature pyramid networks for object detection, in Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, (2017), 2117–2125. https://doi.org/10.1109/CVPR.2017.106
[19]	L. Yang, R. Y. Zhang, L. Li, X. Xie, Simam: A simple, parameter-free attention module for convolutional neural networks, in International Conference on Machine Learning, (2021), 11863–11874.
[20]	R. Sunkara, T. Luo, No more strided convolutions or pooling: A new CNN building block for low-resolution images and small objects, in Machine Learning and Knowledge Discovery in Databases, Springer Nature, Cham, Switzerland, (2023), 443–459. https://doi.org/10.1007/978-3-031-26409-2_27
[21]	Q. Tian, R. Hu, Z. Li, Y. Cai, Z. Yu, Insulator detection based on se-YOLOv5s, Chin. J. Intell. Sci. Technol., 3 (2021), 312–321. https://doi.org/10.11959/j.issn.2096-6652.202132 doi: 10.11959/j.issn.2096-6652.202132
[22]	B. S. Webb, N. T. Dhruv, S. G. Solomon, C. Tailby, P. Lennie, Early and late mechanisms of surround suppression in striate cortex of macaque, J. Neurosci., 25 (2005), 11666–11675. https://doi.org/10.1523/JNEUROSCI.3414-05.2005 doi: 10.1523/JNEUROSCI.3414-05.2005
[23]	J. Hu, L. Shen, G. Sun, Squeeze-and-excitation networks, in Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, (2018), 7132–7141. https://doi.org/10.1109/TPAMI.2019.2913372
[24]	Q. L. Wang, B. G. Wu, P. F. Zhu, P. H. Li, W. M. Zuo, Q. H. Hu, ECA-Net: Efficient channel attention for deep convolutional neural networks, in Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, (2020), 11531–11539. https://doi.org/10.1109/CVPR42600.2020.01155
[25]	S. Woo, J. Park, J. Y. Lee, I. S. Kweon, CBAM: convolutional block attention module, in Proceedings of the European Conference on Computer Vision (ECCV), (2018), 3–19. https://doi.org/10.48550/arXiv.1807.06521
[26]	A. Dewangan, Y. Pande, H. W. Braun, F. Vernon, I. Perez, I. Altintas, et al., Figlib & smokeynet: Dataset and deep learning model for real-time wildland fire smoke detection, Remote Sens., 14 (2022), 1007. https://doi.org/10.3390/rs14041007 doi: 10.3390/rs14041007
[27]	K. Govil, M. L. Welch, J. T. Ball, C. R. Pennypacker, Preliminary results from a wildfire detection system using deep learning on remote camera images, Remote Sens., 12 (2020), 166. https://doi.org/10.3390/rs12010166 doi: 10.3390/rs12010166

Reader Comments

Your name:*

Email:*
© 2023 the Author(s), licensee AIMS Press. This is an open access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/4.0)