Research on Safe Wearable Target Detection Technology for Power Production Scenarios based on YOLOv5-C3CA

Chiyi Ma; Xingzhong Xiong; Jun Liu

doi:10.54691/qayscj09

Authors

Chiyi Ma
Xingzhong Xiong
Jun Liu

DOI:

https://doi.org/10.54691/qayscj09

Keywords:

Target Detection; YOLOv5; Attention Mechanism; AFPN.

Abstract

Aiming at the establishment of a new generation of substation auxiliary equipment detection platform intelligence, this study proposes a power production industry personnel safety wear detection technology based on improved Yolov5-C3CA to meet the demand. First, the network performance is improved by adding the CA attention mechanism module to Yolov5s network; moreover, the CA module is modified to C3CA module and added to the model for more effective addition of the attention mechanism; finally, the AFPN network structure is used to replace the original feature pyramid network structure, which further effectively improves the utilization efficiency of shallow and deep features. The experimental results show that the mean average accuracy of the designed network is improved by 3.9% to 93.1% compared with the original network, and other evaluation indexes are also improved. It can be seen that the model modification in this paper has improved the performance of the detection network, and the improved network meets the requirements of the new generation of substation auxiliary equipment detection platform, which has a positive effect on the safety and reliability of the power production industry.

Downloads

Download data is not yet available.

References

RogersAnna, GardnerMatt, AugensteinIsabelle. QA Dataset Explosion: a Taxonomy of NLP Resources for Question Answering and Reading Comprehension [J]. ACM Computing Surveys, ACM, 2023.

Narayan V, Awasthi S, Fatima N, et al. Deep Learning Approaches for Human Gait Recognition: a Review[A]. 2023 International Conference on Artificial Intelligence and Smart Communication (AISC)[C]. IEEE, 2023: 763-768.

Malhotra P, Gupta S, Koundal D, et al. Deep Neural Networks for Medical Image Segmentation[J]. Journal of Healthcare Engineering, Hindawi, 2022, 2022: e9580991.

Ang L, Rahim S K N A, Hamzah R, et al. YOLO algorithm with hybrid attention feature pyramid network for solder joint defect detection[J]. arXiv, 2024.

Wang H, Yang G, Li E, et al. High-Voltage Power Transmission Tower Detection Based on Faster R-CNN and YOLO-V3[A]. 2019 Chinese Control Conference (CCC)[C]. 2019: 8750-8755.

Kang F, Li J. Research on the Detection Method of Electric Power Workers not Wearing Helmets based on YOLO Algorithm[A]. Proceedings of the 2023 9th International Conference on Computing and Artificial Intelligence[C]. New York, NY, USA: Association for Computing Machinery, 2023: 66-71.

Souza B J, Stefenon S F, Singh G, et al. Hybrid-YOLO for classification of insulators defects in transmission lines based on UAV[J]. International Journal of Electrical Power & Energy Systems, 2023, 148: 108982.

QI Zezheng, XU Yinxia. Research on helmet wearing detection with improved YOLOv5s algorithm[J]. Computer Engineering and Application, 2023: 1-10.

Redmon J, Divvala S, Girshick R, et al. You Only Look Once: Unified, Real-Time Object Detection[A]. 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR)[C]. Las Vegas, NV, USA: IEEE, 2016: 779-788.

Redmon J, Farhadi A. YOLO9000: Better, Faster, Stronger[J]. arXiv, 2016.

Redmon J, Farhadi A. YOLOv3: An Incremental Improvement[J]. : 6.

Bochkovskiy A, Wang C-Y, Liao H-Y M. YOLOv4: Optimal Speed and Accuracy of Object Detection[J]. arXiv, 2020.

He K, Zhang X, Ren S, et al. Spatial Pyramid Pooling in Deep Convolutional Networks for Visual Recognition[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2015, 37(9): 1904-1916.

Liu S, Qi L, Qin H, et al. Path Aggregation Network for Instance Segmentation[A]. 2018: 8759-8768.

Lin T-Y, Dollar P, Girshick R, et al. Feature Pyramid Networks for Object Detection[A]. 2017: 2117-2125.

Hu J, Shen L, Albanie S, et al. Squeeze-and-Excitation Networks[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2020, 42(8): 2011-2023.

Woo S, Park J, Lee J-Y, et al. Cbam: Convolutional block attention module[A]. Proceedings of the European conference on computer vision (ECCV)[C]. 2018: 3-19.

Park J, Woo S, Lee J-Y, et al. BAM: Bottleneck Attention Module[J]. arXiv, 2018.

Liu Y, Shao Z, Hoffmann N. Global Attention Mechanism: Retain Information to Enhance Channel-Spatial Interactions[J]. arXiv, 2021.

Hou Q, Zhou D, Feng J. Coordinate Attention for Efficient Mobile Network Design[A]. 2021: 13713-13722.

Wang Q, Wu B, Zhu P, et al. ECA-Net: Efficient Channel Attention for Deep Convolutional Neural Networks[J]. arXiv, 2020.

Liu S, Qi L, Qin H, et al. Path Aggregation Network for Instance Segmentation[A]. 2018: 8759-8768.

Tan M, Pang R, Le Q V. Efficientdet: scalable and efficient object detection[A]. Proceedings of the IEEE/CVF conference on computer vision and pattern recognition[C]. 2020: 10781-10790.

Yang G, Lei J, Zhu Z, et al. AFPN: Asymptotic Feature Pyramid Network for Object Detection[J]. 2023.

Liu S, Huang D, Wang Y. Learning Spatial Fusion for Single-Shot Object Detection[J]. arXiv, 2019.

Liu W, Anguelov D, Erhan D, et al. SSD: Single Shot MultiBox Detector[A]. B. Leibe, J. Matas, N. Sebe, et al. Computer Vision - ECCV 2016[C]. Cham: Springer International Publishing, 2016: 21-37.