Automated Dress Code Detection System for Construction Site Safety

Abstract

The project presents an innovative approach to construction safety and hazard detection by integrating the YOLOv8 object detection model with the Django web framework. By leveraging the advanced capabilities of YOLOv8, known for its high accuracy and real-time detection performance, the system identifies potential safety hazards on construction sites such as unauthorized personnel, unsafe equipment usage and hazardous materials. The Django framework facilitates seamless integration, allowing for efficient data management, real-time alerts and user-friendly interfaces for monitoring and reporting. The proposed solution enhances proactive safety measures, reducing the risk of accidents and improving overall site safety. Experimental results demonstrate the system\'s effectiveness in detecting a wide range of hazards, showcasing its potential as a critical tool for ensuring safer construction environments.

Introduction

Construction site safety is vital as workers face hazardous conditions, and many accidents result from missing personal protective equipment (PPE). Traditional safety monitoring is manual and prone to errors. This project develops an automated real-time safety monitoring system using YOLOv8 for fast, accurate detection of helmets, masks, and vests, integrated with the Django web framework for data management, alerts, and a user-friendly dashboard.

Building on previous PPE detection studies that used earlier YOLO versions and other machine learning models, this system advances accuracy and deployment ease by combining YOLOv8 with Django and cloud infrastructure. The system includes modules for data preprocessing, real-time detection, alerting, and a chatbot (SafeBot) that provides health assistance for minor injuries.

Testing showed that YOLOv8 achieved around 92% accuracy in detecting safety gear under varied conditions, with rapid alert generation and effective integration for scalable monitoring. The system enhances construction site safety by automating PPE compliance detection and providing timely notifications.

Conclusion

The integration of YOLOv8 with Django enhances construction site safety by enabling real-time hazard detection and management. YOLOv8’s robust object detection capabilities allow for accurate identification of potential risks, while Django provides a user-friendly web interface for monitoring safety alerts. The proposed system provides a smart and effective way to monitor construction safety in real time. By using YOLOv8 for object detection and Django for deployment, we ensure fast and accurate responses.

References

[1] Z. Chen and Z. Chen, “RBNet: A deep neural network for unified road and road boundary detection,” in Proc. Int. Conf. Neural Inf. Process., 2017 , pp. 677–687. [2] J. Munoz-Bulnes, C. Fernandez, I. Parra, D. Fernández-Llorca, and M. A. Sotelo, “Deep fully convolutional networks with random data augmentation for enhanced generalization in road detection,” in Proc. IEEE Int. Conf. Intell. Transp. Syst., 2017, pp. 366–371. [3] R. Liu, Z. Yuan, T. Liu, and Z. Xiong, “End-to-end lane shape prediction with transformers,” in Proc. IEEE Winter Conf. Appl. Comput. Vis., 2021 , pp. 3694–3702. [4] Z. Qin, H. Wang, and X. Li, “Ultra fast structure-aware deep lane detection,” in Proc. Eur. Conf. Comput. Vis., 2020, pp. 276–291. [5] S. Ren, K. He, R. Girshick, and J. Sun, “Faster R-CNN: Towards real-time object detection with region proposal networks,” in Proc. Int. Conf. Neural Inf. Process. Syst., 2015, pp. 91–99. [6] P. Dollár, C. Wojek, B. Schiele, and P. Perona, “Pedestrian detection: An evaluation of the state of the art,” IEEE Trans. Pattern Anal. Mach. Intell., vol. 34, no. 4, pp. 743–761, Apr. 2012. [7] F. Chabot, M. Chaouch, J. Rabarisoa, C. Teuliere, and T. Chateau, “Deep MANTA: A coarse-to-fine many-task network for joint 2D and 3D vehicle analysisfrommonocularimage,”inProc.IEEEConf.Comput.Vis.Pattern Recognit., 2017, pp. 1827–1836. [8] H. Zhao, J. Shi, X. Qi, X. Wang, and J. Jia, “Pyramid scene parsing network,” in Proc. IEEE Conf. Comput. Vis. Pattern Recognit., 2017 , pp. 6230–6239. [9] L.-C. Chen, G. Papandreou, F. Schroff, and H. Adam, “Rethinking atrous convolution for semantic image segmentation,” 2017, arXiv: 1706.05587. [10] E. Romera, J. M. Alvarez, L. M. Bergasa, and R. Arroyo, “ERFNet: Efficient residual factorized ConvNet for real-time semantic segmentation,” IEEE Trans. Intell. Transp. Syst., vol. 19, no. 1, pp. 263 272, Jan. 2018.

Copyright

Copyright © 2025 Samitha T, Shareen Fathima S, Praveen K V. This is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.