The system of detecting safety helmets based on YOLOv5
Guanhao Yang, Qujiang Lei · 2021 International Conference on Electronic Information Engineering and Computer Science (EIECS) · 2021
The rapid development of computer vision is applied in various aspects and has a wide prospect, especially in safety engineering. In construction sites, the helmet is an important tool to protect workers' life, but actually, because of not wearing a helmet, the accident happened sporadically. To solve this problem, based on deep learning, we collected 2241 images for training and testing, additionally, using YOLOV5 to recognize some common images. The experimental results are shown in the proposed method, and the average accuracy can reach about 95.2%.