Smart Personal Protective Equipment (PPE) Systems: Integrating Deep Learning to Improve Workplace Safety

The use of PPE is mandatory for workers in organizations with high risks, such as construction companies and factories. However, sustaining the correct and appropriate use of PPE is one of the significant concerns in healthcare organizations. This study proposes the design and test of a SMART PPE system that incorporates deep learning to improve safety. The approach described in the proposed system utilizes the use of real-time object detection coupled with sensor data to monitor and detect the presence as well as usage of PPEs and, in case of non-compliance, alert the workers as well as supervisors. The present paper discusses the development of a SMART PPE system with deep learning to minimize the risk of workplace hazards. To detect compliance with PPEs, real-time object detection with YOLOv7 is used, and the maximum accuracy of 97. 29%, precision of 91. 75%, recall of 96. 39%, and F1 score of 94. 01%. Through the inclusion of sensors and edge computing, the system gives an automated and real-time alert system and minimizes human supervision to improve safety measures. The study also focuses on the performance of the system and discusses possible problems, which are the reliability of the sensors and the requirement for constant connection, which are suggested in a number of directions in future research.