Detection of dress code for anti-static equipment is an important management link in clean workshops. To address the issue of difficulty in deploying multi-scale dress code detection methods for anti-static equipment in embedded systems, a lightweight real-time detection method for dress code of anti-static equipment is proposed. This article uses the MobileNetV3-small backbone network to extract features of anti-static equipment, making the model lightweight and easy to deploy. Adopting BiFPN structure to enhance the feature fusion ability of anti-static equipment at multiple scales, and using CIoU Loss and DIoU-NMS to accurately locate anti-static equipment targets, and improving the problem of missed detection of anti-static equipment when people are crowded, and improving the accuracy of dress code detection for anti-static equipment. The experimental results show that the algorithm improves accuracy by 2.1%, reduces parameter count by 43.8%, and reduces model size by 40.6% compared to YOLOv5s. The recognition speed on the Jeston Xavier NX system is 27FPS, and the recognition accuracy of wearing anti-static hats, anti-static clothing, and anti-static shoes is 98.1%, 96.2%, 95.8%, 94.2%, and 94.1%, respectively. It meets the requirements of real-time detection of anti-static equipment dress code.

Qi Li, Jin Wei. Lightweight Real-time Detection Method for Dress Code of Anti-static Equipment. Academic Journal of Computing & Information Science (2023), Vol. 6, Issue 10: 7-17. https://doi.org/10.25236/AJCIS.2023.061002.

[1] Li S X, Fan Z K, Liu L. Interpretation of GB 12014-2019 “Protective Clothing and Antistatic Clothing” [J]. China Personal Protective Equipment, 2020 (Z2): 26-31. 

[2] Chen S L. Anti-static work clothes and their testing standards [J]. China Labor Protection Equipment, 1996 (02): 14-19. 

[3] Zhao L R. Research on deep learning algorithm for helmet wearing detection under video surveillance [D]. Beijing: North University of Technology, 2022:1-9. 

[4] Li J Q, Zhao X F, Zhou G Y, et al. Standardized use inspection of workers' personal protective equipment based on deep learning[J]. Safety Science, 2022, 150. 

[5] Nath, N. D., Behzadan, A. H., Paal, S. G., Deep learning for site safety: Real-time detection of personal protective equipment. Autom. Constr. 2020. 112, 103085. 

[6] Chen, S., Demachi, K. Towards on-site hazards identification of improper use of personal protective equipment using deep learning-based geometric relationships and hierarchical scene graph. Autom. Constr. 2021. 125, 103619. 

[7] Ding P, Kulban A, Geng L T, et al. Real time facial mask detection and standardized wearing recognition in natural environments [J]. Computer Engineering and Applications, 2021, 57 (24): 268-275. 

[8] He L, Ge X, Hao C Y, et al. Identification of Dress Code for Substation Operators Based on YOLOv5 [J]. Power Big Data, 2021, 24 (10): 1-8. 

[9] Mao F, Ling Y B, Guo Y, et al. Detection of electrician dress irregularities based on multi-scale attention networks [J]. Computer Knowledge and Technology, 2022, 18 (21): 4-7+14. 

[10] Howard A, Sandler M, Chen B, et al. Searching for MobileNetV3 [C]//2019 IEEE/ CVF International Conference on Computer Vision (ICCV), 2019:1314-1324. 

[11] Wang Z P, Zhang R F, Liu Y H, etc. An improved YOLOv3 garbage classification detection model for edge computing devices [J]. Progress in Laser and Optoelectronics, 2022, 59 (04): 291-300. 

[12] Zhu X, Lyu S, Wang X, et al. TPH-YOLOv5: Improved YOLOv5 based on transformer prediction head for object detection on drone-captured scenarios[C]//Proceedings of the IEEE/CVF International Conference on Computer Vision. 2021: 2778-2788. 

[13] Wu T S, Zhang Z J, Liu Y P, et al. Driver Seat Belt Detection Based on YOLO Detection and Semantic Segmentation [J]. Journal of Computer Aided Design and Graphics, 2019, 31 (01): 126-131. 

[14] Tan M, Pang R, Le Q-V. Efficient Det: Scalable and Efficient Object Detection[C]//IEEE Conference on Computer Vision and Pattern Recognition (CVPR), Seattle, USA: IEEE, 2020: 10778-10787. 

[15] Zheng Z H, Wang P, Liu W, et al. Distance-IoU Loss: Faster and better learning for bounding box begression [J]. arXiv preprint arXiv:1911. 08287, 2019. 

[16] Yang Y B, Li D. Improving the Lightweight Helmet Wearing Detection Algorithm for YOLOv5 [J]. Computer Engineering and Applications, 2022, 58 (09): 201-207. 

[17] Iu W, Anguelov D, Erhan D, et al. SSD: single shot multibox detector [C]//European Conference on Computer Vision. Cham: Springer, 2016: 21-37. 

[18] Ren S Q, He K M, Girshick R, et al. Faster R-CNN: Towards real-time object detection with region proposal Networks [J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2017, 39(6):1137-1149.