A Systematic Review of Helmet Wearing Detecting Algorithms Based on YOLO

In industrial production and social transportation, there are often various potential risks that threaten the life safety of participants. In all workers' lives in the most serious, generally likely to happen are falling objects, collapses, and traffic accidents, etc. These serious lessons let all workers live on the helmet wearing and supervision more and more attention, effective supervision will reduce accidents. Therefore, we collected literature and provided an overview of deep learning algorithms related to helmet detection. After comparing features of popular object detection algorithms, YOLO algorithm was chosen as the research object in this field. First, the development history of the YOLO algorithm (YOLOvl∼YOLOv8) is described, and then the literature on YOLO-based helmet wearing detection is reviewed. Second, we analyze selected literature to identify current knowledge, key gaps, and improvements in YOLO for specific helmet wearing detection tasks. Finally, future improvement ideas and research trends are analyzed and prospected.