Design of a Personal Protective Equipment detection system using Computer Vision and Convolutional Neural Networks

Villagomez B.K.M.; Huachaca P.A.Q.; Suarez J.T. · SSRG International Journal of Electronics and Communication Engineering (2025) · DOI: 10.14445/23488549/ijece-v12i4p115

Resumen

In work environments, the identification and correct use of Personal Protective Equipment (PPE) is essential to avoid incidents and safeguard employees' health. In this paper, an innovative strategy for the automatic identification of PPE using computer vision techniques and Convolutional Neural Networks (CNN) is presented. The method uses a specially trained CNN to interpret images of the equipment and a labeled dataset that was developed especially for PPE detection. Standard item detection criteria were used to evaluate the system's performance, and they were shown to be effective in correctly identifying Personal Protective Equipment (PPE) in photographs of industrial environments. The results of this study show a great degree of sensitivity and accuracy in the identification of several kinds of Personal Protective Equipment (PPE), indicating that the development of this technology can improve automated inspection tasks and safety in the industrial workplace, avoiding dangerous circumstances by providing better control.

Instituciones: Universidad Nacional de San Agustín de Arequipa