The Frontiers of Society, Science and Technology, 2025, 7(2); doi: 10.25236/FSST.2025.070203.
Pengfei Ma1, Minghao Cao2, Chang Cai3, Bo Fang4
1School of Electronic Information, Xijing University, Xi’an, China
2School of Electronic Information, Xijing University, Xi’an, China
3School of Electronic Information, Xijing University, Xi’an, China
4School of Electronic Information, Xijing University, Xi’an, China
With the increasing emphasis on safety management in industrial settings, ensuring that power maintenance workers wear proper protective equipment has become a critical concern. Traditional safety monitoring methods often rely heavily on manual intervention, leading to inefficiencies and susceptibility to human error. To address these challenges, computer vision and deep learning technologies provide an effective solution for automated safety detection. This study employs an improved YOLO (You Only Look Once) algorithm to enhance detection accuracy while optimizing efficiency and real-time performance. The proposed approach meets the safety management requirements of power maintenance operations by enabling intelligent monitoring of workers' protective equipment. By automating safety inspections, the system significantly reduces the risk of accidents and enhances worker safety, contributing to a more secure and efficient maintenance environment.
Safety Protection Detection, Computer Vision, Deep Learning
Pengfei Ma, Minghao Cao, Chang Cai, Bo Fang. The Application of YOLOv10 Model Integrated with RepViT in Safety Protective Equipment Detection for Power Maintenance Workers. The Frontiers of Society, Science and Technology (2025), Vol. 7, Issue 2: 16-22. https://doi.org/10.25236/FSST.2025.070203.
[1] Zell O, Pålsson J, Hernandez-Diaz K, et al. Image-Based Fire Detection in Industrial Environments with YOLOv4[J]. arXiv preprint arXiv:2212.04786, 2022.
[2] Islam M S, Shaqib S M, Ramit S S, et al. A Deep Learning Approach to Detect Complete Safety Equipment For Construction Workers Based On YOLOv7[J]. arXiv preprint arXiv:2406.07707, 2024.
[3] Chen Q, Su X, Zhang X, et al. LW-DETR: A Transformer Replacement to YOLO for Real-Time Detection[J]. arXiv preprint arXiv:2406.03459, 2024.
[4] Hu J, Shen L, Sun G. Squeeze-and-Excitation Networks[C]. Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, 2018: 7132-7141.
[5] Amini A, Shirani-Mehr H, Karbasi A. Digital Modulation Classification: A Deep Learning Approach[J]. IEEE Transactions on Communications, 2017, 65(11): 4658-4668.
[6] He K, Zhang X, Ren S, et al. Deep Residual Learning for Image Recognition[C]. Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, 2016: 770-778.
[7] Redmon J, Divvala S, Girshick R, et al. You Only Look Once: Unified, Real-Time Object Detection[J]. arXiv preprint arXiv:1506.02640, 2015.
[8] Çiftçi M, Türkdamar M U, Öztürk C. Leveraging YOLO Models for Safety Equipment Detection on Construction Sites[J]. Journal of Computing Theories and Applications, 2024, 1(1): 1-8.
[9] Zhang Z, Li H, Zhang L. Parallel Personal Protective Equipment Recognition Network Based On Multimodal Fusion[C]. Proceedings of the 31st ACM International Conference on Multimedia, 2023: 1234-1242.
[10] Hussain M. YOLOv5, YOLOv8 and YOLOv10: The Go-To Detectors for Real-time Vision[J]. arXiv preprint arXiv:2407.02988, 2024.
[11] Lu Z, Li Y, Wang C, et al. YOLOv5-RT: A Lightweight Real-Time YOLOv5 Model for Edge Devices[J]. arXiv preprint arXiv:2303.08665, 2023.
[12] Wang A, Chen H, Lin Z, et al. RepViT: Revisiting Mobile CNN From ViT Perspective[J]. arXiv preprint arXiv:2307.09283, 2023.
[13] Laakom F, Chumachenko K, Raitoharju J, et al. Learning to Ignore: Rethinking Attention in CNNs[J]. arXiv preprint arXiv:2111.05684, 2021.