Research on improved YOLOv5 face mask recognition algorithm based on
Yan Zhang, Ying Cuan, Yuxin Jiang · 2023
Accurately detecting whether a person is wearing a mask or not is important for preventing the spread of infectious diseases. In this paper, we propose a research method based on the improved YOLOv5 face mask recognition, which combines SE, CA and CBAMc3 attention mechanisms with YOLOv5 model for face mask recognition problem, and compare the effects of different attention mechanisms on recognition accuracy. The experimental results show that combining the SE attention mechanism with YOLOv5 can significantly improve the face mask recognition accuracy, and the mAP of the model is improved by 1.3%, while the CA and CBAMc3 mechanisms are less effective on the accuracy improvement. This study explores the effectiveness of different attention mechanisms in face mask recognition, and provides new ideas and methods for research in this field.