摘要
本文将表情识别技术应用于课堂教学,建立了一个基于面部表情的学生课堂专注度评价系统,可以给教师提供更加准确和实时的教学评价。首先使用局部二值模式(Local binary patterns,LBP)和局部相位量化(Local Phase Quantization,LPQ)提取图像特征,其次将其融合成为一个新特征并送入支持向量机(Support Vector Machine,SVM)分类器进行表情识别,最后计算专注度得分,并进行学生课堂专注度评价。通过实验,实验结果证明使用该算法进行学生课堂专注度评价是相对可行的,可以为教师的教学评价提供一个客观公正的依据。
In this paper, we apply expression recognition technology to classroom teaching and build a classroom concentration evaluation system based on facial expressions, which can provide teachers with more accurate and real-time teaching evaluation. Firstly,the image features are extracted using Local binary patterns(LBP) and Local Phase Quantization(LPQ), secondly, they are fused into a new feature and fed into an Support Vector Machine(SVM) classifier for expression recognition, and finally, the concentration score is calculated and the students’ classroom concentration is evaluated. The experimental results demonstrate that using this algorithm for student classroom concentration evaluation is relatively feasible and can provide an objective and fair basis for teachers’ teaching evaluation.
作者
韦宁燕
韦洪浪
WEI Ningyan;WEI Honglang(Nanning College of Technology,Guilin Guangxi 541006,China;Guilin University of Technology at Nanning,Nanning Guangxi 530001,China)
出处
《信息与电脑》
2022年第21期176-179,共4页
Information & Computer
基金
2021年度广西高校中青年教师科研基础能力提升项目“基于深度学习的学生课堂表情识别与课堂专注度分析的研究”(项目编号:2021KY1671)
2020年度广西高等教育本科教学改革工程项目“基于创新创业的《大学物理》课程教学改革与实践——基于雏形教具的应用”(项目编号:2020JGB477)。
