基于YOLOX模型的口罩目标检测研究与应用  

Research and Application of Mask Object Detection Based on YOLOX Model

作　　者：郭永跃 于洋[2] GUO Yongyue;YU Yang(School of Software,Tianjin Normal University,Tianjin 300387,China;School of Computer and Information Engineering,Tianjin Normal University,Tianjin 300387,China)

机构地区：[1]天津师范大学软件学院,天津300387 [2]天津师范大学计算机与信息工程学院,天津300387

出　　处：《微型电脑应用》2024年第3期93-96,共4页Microcomputer Applications

基　　金：天津师范大学校级教学改革项目(JGYB01218046);天津师范大学校青年基金(52XQ1403)。

摘　　要：针对公共场所存在人员密集、异物遮挡面部以及不规范佩戴口罩等问题,以YOLOX网络为基础对人们是否佩戴口罩进行识别,同时使用GIoU损失函数,完善模型的检测效果;使用Albumentations库进行数据增强,提高捕捉面部的准确率;添加GUI设计,使用户的操作更加便捷。实验结果表明,该模型的平均准确率(mAP)达到了94.36%,可以在光线昏暗等各种环境因素下和面部存在各种遮挡物下准确识别口罩是否遮住口鼻,实时性较好,未来可在教室、商场等公共场所用于检测。In view of the problems of crowded people,foreign bodies may block faces and non-standard wearing masks in public places,YOLOX network is used to identify whether people wear masks,and GIoU loss function is used to improve the detection effect of the model.Albumentations library is used for data enhancement,which can improve the accuracy of capturing faces;GUI design is added to make the user's operation more convenient.The experimental results show that the average accuracy(mAP)of the model reaches 94.36%,which can accurately identify whether the mask covers the nose and mouth under various environmental factors such as dim light and various obstructions on the face.At the same time,the model has terrific real-time performance and it can be used for testing in classrooms,mall and other public in the future.

关 键 词：规范佩戴口罩 YOLOX网络 GIoU GUI设计 

分 类 号：TP391.41[自动化与计算机技术—计算机应用技术]