基于深度学习的线上教学学情监测系统研究

Study on Monitoring System of Learning Situation in Online Teaching Based on Deep Learning

线上教学在疫情期间得到广泛应用,但是在授课过程中,教师没有精力关注每一位学生对讲授内容的实时反应,无法针对性地调整授课方案,导致教学效果不佳。因此,本文开发了基于深度学习的线上教学学情监测系统。该系统采用Yolov3算法,结合摄像头实时采集的视频画面,对学生进行检测定位,计算学生在线时间。同时,其使用Dlib开源库检测学生面部特征点,对学生进行困倦状态识别,并生成课堂学情报告,分析学生出勤、学习状态数据。测试表明,该系统响应迅速,可靠性较高,能准确判断学生的学习状态,具有较好的实际应用价值。

Online teaching has been widely used during the epidemic,but in the course of teaching,teachers did not have the energy to pay attention to the real-time response of each student to the content of the lecture,and could not adjust the teaching plan in a targeted manner,resulting in poor student learning.Therefore,this paper developed an online education monitoring system based on deep learning.The system uses the Yolov3 algorithm,combined with the real-time video images collected by the camera,detects and locates students,and calculates the online time of students.At the same time,it uses the Dlib open source library to detect student facial feature points,recognize students'sleepiness status,and generate classroom learning reports,and analyze all students'absenteeism and learning status data.Tests show that the system responds quickly,has high reliability,can accurately judge the learning status of students,and has high practical application value.