摘要
面部表情是表达情绪、情感的直接方式,掌握课堂教学中学生的面部表情信息,有助于教师多途径掌控课堂情况。本研究通过深度学习模型提取学生课堂面部表情图像的关键特征信息,从而识别课堂中学生高兴、愤怒、平常、困惑、惊讶、悲伤、恐惧、厌恶时的表情。结果表明:基于深度学习的学生课堂表情识别方法可有效排除教室背景、穿着等无关信息的干扰,对突出相关信息具有较好的效果。使用基于深度学习的学生课堂面部表情识别方法能及时有效反映学生的学习状态,有利于帮助教师察觉学生情绪,掌握学生的课堂学情,改善课堂的过程性评价,从而助力课堂教学高质量发展。
Facial expressions are the most direct and effective way to express emotions and emotions.Obtaining and mastering the facial expression information of students in classroom teaching is of great significance for teachers to control the classroom situation through multiple channels.In this study,the deep learning model was used to extract the key feature information of students'facial expression images in class,so as to identify the expressions of students in class when they are hAppy,angry,normal,confused,surprised,sad,afraid and disgusted.The results show that the method of recognizing students,classroom expressions can effectively eliminate the interference of irrelevant information such as classroom background and students'clothing,and has a good performance on highlighting relevant information.The use of deep-leaming-based students,facial expression recognition method in the classroom is conducive to detecting students,emotions,improving the classroom teaching evaluation mechanism,timely and effectively reflecting student'learning status,and helping teachers accurately grasp students,classroom learning conditions,thereby helping the high-quality development of classroom teaching.
作者
牟昱睿
雷紫珺
田肖宜
刘香一
MOU Yurui;LEI Zijun;TIAN Xiaoyi;LIU Xiangyi
出处
《湖南广播电视大学学报》
2023年第1期17-24,共8页
Journal of Hunan Radio and Television University
基金
广西教育科学“十四五”规划广西教育信息化发展研究委托重点课题“民族地区教育信息化融合创新的实践研究”(2022AA11)
国家社会科学基金教育学青年课题“教育变革的文化基因研究”(CAA200237)。
关键词
深度学习
表情识别
课堂评价
deep learning
facial expressions
classroom assessment
