Mobile mask detection system based on YOLOv4-tinys

Wearing a mask is an effective way to prevent the spread of disease, especially in crowded public areas, and it is essential to keep track of how well people are wearing masks. It would be very costly to dispatch people to monitor the wearing of masks. This paper takes advantage of the development of computer vision technology in image recognition to apply image recognition technology to mask detection to monitor the wearing of masks automatically and accurately. It is building a mask detection model based on YOLOv4-tiny and quantifying them at a high level of accuracy. The model is converted to Tensorflow Lite format and then installed on the Android device. It realises the accuracy and real-time of mask-wearing detection, enhances the system’s ease of use and portability, and has important practical promotion value.