Facial Mask-Wearing Detection based on Deep Learning

It is particularly important to urge people to wear masks in the scene with a large flow of people. Previously, nondeep learning mask detection methods mainly relied on image processing technology, such as color feature and texture feature analysis based on traditional computer vision. However, these methods are susceptible to factors like lighting and occlusion. Meanwhile, the detection accuracy is limited under diverse facial postures and the generalization ability is weak. Therefore, developing an efficient intelligent mask-wearing detection system has important practical significance. In response to the above problems, this paper proposes a face mask-wearing detection method based on a multi-level convolutional neural network. The experiment uses public data sets to train the network. The multilevel convolutional neural network automatically learns the deep features of human faces and masks, thereby improving detection accuracy. The experimental results show that the proposed method is superior to the traditional method in terms of indicators such as accuracy rate. It effectively achieves intelligent detection, providing a feasible technical solution for the supervision of maskwearing in public places and demonstrating broad prospects in practical applications.