Published Paper Details:

Construction had the highest number of fatal work accidents of any industry, due to the high number of accidents each year. There are many solutions to ensure worker safety and reduce accidents, one of which is to ensure the proper use of appropriate personal protective equipment (PPE kit) as defined in safety regulations. However, monitoring the use of personal protective equipment, which is largely based on manual checks, is time-consuming and inefficient. The Several attempts were made. The resulting recognition accuracy on the 12 main armors is up to 98%, while the accuracy of face detection and recognition is 96%. The obtained results showed the ability to identify recognize faces and all the equipment very accurately and remember them in real time. The study uses convolutional neural network (CNN) models developed by applying transfer learning to the basic version of YOLOv7 and The Faster RCNN deep learning. Considering the presence the model predicts fulfillment of requirements in 12 categories, such as jacket, jacket, mask, no mask, shoes, no shoes, helmet, no helmet, Safety Belt, No Safety Belt, No front Safety Belt, Front Safety Belt.