Advancing Safety Protocols with Deep Learning: An Efficient Helmet Detection System

The safety of offshore workers in the petroleum industry is a paramount concern, especially in unpredictable and unsafe conditions that need to be managed in real time due to the nature of the industry. This study describes a helmet detection system through a deep learning method for demonstrating enhanced safety compliance in an industrial environment. The system utilizes advanced convolutional neural networks (CNNs) to detect and classify the helmets worn by workers in implantation through surveillance. The dataset had various conditions of the environment, differences in pose, and lighting that were used to develop and validate the helmet detection model. The users' experimental study reports high precision with high recall and is appropriate for application in large deployment scenarios to monitor safety against the use of protective equipment. The application of YOLOv۱۲, a novel object detection framework, to petroleum engineering is a valuable and innovative use of deep learning in a niche industrial domain.

Keywords:

Helmet Detection , Worker Safety , Object Detection , Computer Vision