摘要
传统的课堂教学评价往往效率低下,并带有较强的主观性。针对传统课堂评价中存在的不足,结合深度学习技术,在CNN模型基础上建立起适合课堂场景的人脸检测和表情识别模型,得到比较准确的人脸特征,接着使用朴素贝叶斯分类器对得到的人脸特征进行分类和评价,然后研究面部特征与课堂质量之间的关系,最后建立起基于人脸检测和表情识别的课堂评价规则。实验数据表明,本研究可以作为课堂教学评价的重要参考指标。
Traditional classroom teaching evaluation is often inefficient and subjective. In terms of the shortcomings existing in traditional classroom evaluation, combined with the technology of deep learning, face detection and facial expression recognition model suitable for classroom scene was established on the basis of the model of CNN and more accurate facial feature was obtained. The naive bayesian classifier was then used for face feature classification and evaluation, and the relationship between facial features and classroom quality was researched. Classroom evaluation rules were finally built up on the basis of face detection and facial expression recognition. Experimental data show that this study can be used as an important reference index for classroom teaching evaluation.
作者
梁利亭
LIANG Li-ting(Sanmenxia Polytechnic,Sanmenxia,Henan 472000,China)
出处
《晋城职业技术学院学报》
2020年第2期40-44,共5页
Journal of Jincheng Institute of Technology
基金
河南省大中专院校就业创业2018年度课题《“互联网+”时代下智慧校园服务平台研究》(项目编号:JYB2018305)。
关键词
人脸检测
表情识别
课堂教学评价
face detection
facial expression recognition
classroom teaching evaluation
