基于目标检测的口罩识别系统的设计与实现

中南财经政法大学201N届本科生毕业论文（设计）AbstractWith the rapid development of deep learning and computer vision,related technologies have been widely used in many fields.As an important part of image understanding,object detectionis one of the core problems in the field of machine vision.Its task is to find out all the interestedobjects in the image and determine their position and size.Object detection is an important basicresearch in the field of computer vision.It is the technology of object classification and location.It is an important basic technology of image content understanding.In 2020,novel coronavirus pneumonia novel coronavirus pneumonia swept the world.NewCoronavirus can spread through respiratory droplets.Wearing masks correctly can effectively cutoff the transmission of new crown pneumonia virus,and is an effective measure to prevent infection.It is required to wear masks in domestic public places.However,the response measures of people intensive places such as shopping malls,restaurants and subways during the peak period ofpeople flow often make people not pay attention to it.Moreover,complex factors such as occlusion,dense crowd and small-scale targets often exist in natural scenes,which have an impact on the detection effect of wearing masks.To solve this problem,this system proposes a mask wearing detection algorithm in complexscenes,which can automatically,accurately and quickly identify whether the person in the picture or video is wearing a mask.It has important application scenarios and market prospects.The system adopts a kind of yolo-v4 algorithm based on one stage target detection algorithm of deep learning.Compared with the ordinary face detection model,there is only one category of face,and the face mask detection adds one category,and becomes two categories:the face with mask and the face without mask.Yolo-v4 used in this system is the latest robust target detection technology.Firstly,the image to be detected is extracted and classified hierarchically,and then the target is detected to judge whether the target is 0 or 1,that is,whether the mask is worn or not.Key words:Face recognition;Mask recognition;target detection;YOLO