Wearing a mask is the requirements of many places and environments when entering.When mask face detection is performed manually,it is time-consuming and has a high rate of missed and false detections.Therefore,it is very important to design a model for detecting whether wearing a mask.And the model needs to be lightweight,fast,and highly accurate in order to be applied to real-time video detection.First of all,the traditional MobileNet V2 model is modified by reducing the depth of the model to improve its computational speed.Then a facial database of 1600 face images is built and labeled manually.Finally,the improved model is trained by using the facial database,which is only 11.5 MB.The detection accuracy,loss value,and detection speed of this model were compared with the other three mask detection models through experiments.The average detection accuracy of the detection model proposed in this paper for faces wearing masks has been improved by more than 3%,and the detection speed is also faster than the other three types.
［4］CHENG D，GONG Y H，ZHOU S P，et al.Person Re-Identification by Multi-Channel Parts-Based CNN with Improved Triplet Loss Function［C］//2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR).Las Vegas:IEEE,2016:1335-1344.
