Face detection and mask detection based on convolutional neural network

ABSTRACT

In the covid-19 pandemic is not completely over today, wearing masks travel is still a necessary means to reduce the risk of transmission of the virus. To solve the problem of low efficiency and low accuracy in face mask detection, a convolutional neural network detection model is proposed in this paper. The paper first gives a brief introduction to the basic principles of convolutional neural network, and then builds on the MTCNN network structure, combining the popular Python language, Keras and tensorflow-gpu, this paper constructs a reasonable and effective neural network model for face detection, and then completes the detection of whether a face wears a mask, finally, the simulation results show that the proposed model has good performance and robustness in the presence of occlusion, different angles, dim light and different expressions. © 2022 IEEE.