{"title":"基于深度学习的大学生课堂行为认知与认同","authors":"Xing Su, Wei Wang","doi":"10.5573/ieiespc.2023.12.5.398","DOIUrl":null,"url":null,"abstract":"Recognizing and managing college students\" classroom behavior in a timely manner is of great help in improving teaching quality and strengthening classroom management. This paper builds a model based on the You Only Look Once Version 5 Small (YOLO v5s) algorithm using deep learning to detect and identify college students\" classroom behaviors. The LabelImg annotation tool was used to process the dataset images, and the labeled dataset was the input for the object detection model to recognize college students\" classroom behaviors. Although the precision, recall, mean average precision (mAP), and detection speed of the YOLO v5s model were slightly lower with large classroom densities, compared to medium classroom densities, the difference was negligible. At the same time, the mAP values of the proposed model under three different intersection-over-union thresholds were higher than the single shot multibox detector and regionbased convolutional neural network models, reaching 95.8, 94.3, and 92.9. This paper proves that YOLO v5s can effectively and accurately recognize classroom behavior in real time.","PeriodicalId":37326,"journal":{"name":"IEIE Transactions on Smart Processing and Computing","volume":"17 1","pages":"0"},"PeriodicalIF":0.0000,"publicationDate":"2023-10-31","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":"{\"title\":\"Recognition and Identification of College Students\\\" Classroom Behaviors through Deep Learning\",\"authors\":\"Xing Su, Wei Wang\",\"doi\":\"10.5573/ieiespc.2023.12.5.398\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"Recognizing and managing college students\\\" classroom behavior in a timely manner is of great help in improving teaching quality and strengthening classroom management. This paper builds a model based on the You Only Look Once Version 5 Small (YOLO v5s) algorithm using deep learning to detect and identify college students\\\" classroom behaviors. The LabelImg annotation tool was used to process the dataset images, and the labeled dataset was the input for the object detection model to recognize college students\\\" classroom behaviors. Although the precision, recall, mean average precision (mAP), and detection speed of the YOLO v5s model were slightly lower with large classroom densities, compared to medium classroom densities, the difference was negligible. At the same time, the mAP values of the proposed model under three different intersection-over-union thresholds were higher than the single shot multibox detector and regionbased convolutional neural network models, reaching 95.8, 94.3, and 92.9. This paper proves that YOLO v5s can effectively and accurately recognize classroom behavior in real time.\",\"PeriodicalId\":37326,\"journal\":{\"name\":\"IEIE Transactions on Smart Processing and Computing\",\"volume\":\"17 1\",\"pages\":\"0\"},\"PeriodicalIF\":0.0000,\"publicationDate\":\"2023-10-31\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"0\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"IEIE Transactions on Smart Processing and Computing\",\"FirstCategoryId\":\"1085\",\"ListUrlMain\":\"https://doi.org/10.5573/ieiespc.2023.12.5.398\",\"RegionNum\":0,\"RegionCategory\":null,\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"Q4\",\"JCRName\":\"Engineering\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"IEIE Transactions on Smart Processing and Computing","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.5573/ieiespc.2023.12.5.398","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q4","JCRName":"Engineering","Score":null,"Total":0}
引用次数: 0
摘要
及时认识和管理大学生课堂行为,对提高教学质量、加强课堂管理具有重要意义。本文基于You Only Look Once Version 5 Small (YOLO v5s)算法,利用深度学习技术构建模型,对大学生课堂行为进行检测和识别。使用LabelImg标注工具对数据集图像进行处理,标记后的数据集作为目标检测模型的输入,对大学生课堂行为进行识别。虽然在教室密度较大时,YOLO v5s模型的准确率、召回率、平均平均精度(mAP)和检测速度略低,但与中等教室密度相比,差异可以忽略不计。同时,该模型在三种不同交集-过并阈值下的mAP值均高于单次多盒检测器和基于区域的卷积神经网络模型,分别达到95.8、94.3和92.9。本文证明了YOLO v5s能够有效、准确地实时识别课堂行为。
Recognition and Identification of College Students" Classroom Behaviors through Deep Learning
Recognizing and managing college students" classroom behavior in a timely manner is of great help in improving teaching quality and strengthening classroom management. This paper builds a model based on the You Only Look Once Version 5 Small (YOLO v5s) algorithm using deep learning to detect and identify college students" classroom behaviors. The LabelImg annotation tool was used to process the dataset images, and the labeled dataset was the input for the object detection model to recognize college students" classroom behaviors. Although the precision, recall, mean average precision (mAP), and detection speed of the YOLO v5s model were slightly lower with large classroom densities, compared to medium classroom densities, the difference was negligible. At the same time, the mAP values of the proposed model under three different intersection-over-union thresholds were higher than the single shot multibox detector and regionbased convolutional neural network models, reaching 95.8, 94.3, and 92.9. This paper proves that YOLO v5s can effectively and accurately recognize classroom behavior in real time.
期刊介绍:
IEIE Transactions on Smart Processing & Computing (IEIE SPC) is a regular academic journal published by the IEIE (Institute of Electronics and Information Engineers). This journal is published bimonthly (the end of February, April, June, August, October, and December). The topics of the new journal include smart signal processing, smart wireless communications, and smart computing. Since all electronic devices have become human brain-like, signal processing, wireless communications, and computing are required to be smarter than traditional systems. Additionally, electronic computing devices have become smaller, and more mobile. Thus, we call for papers sharing the results of the state-of-art research in various fields of interest. In order to quickly disseminate new technologies and ideas for the smart signal processing, wireless communications, and computing, we publish our journal online only. Our most important aim is to publish the accepted papers quickly after receiving the manuscript. Our journal consists of regular and special issue papers. The papers are strictly peer-reviewed. Both theoretical and practical contributions are encouraged for our Transactions.