基于“专注”与“走神”表情识别的线上课堂学生专注度评价研究

Research on evaluation of students5 concentration in online classroom based on recognition of"concentration"and"mind-wandering"expressions

线上教学环境中师生之间的情感沟通是真空地带,但师生之间的情感交流是影响教学效果的重要因素。通常而言,“专注”与“走神”是两种难以分辨的学生课堂表情。对线上教学环境中学生“专注”与“走神”两类近似表情的识别,有助于教师对学生的听课状态及时准确地感知。文章提出了一种基于Dlib眼部关键点检测的学生“专注”与“走神”表情识别方法,并在此基础上建立学生表情持续时间统计算法,计算各表情持续的时间。实验结果表明,采用眼部关键点实现上述两种表情识别的准确率为95.1%,且该方法数据量小、计算速度快,结果分析与持续时间统计分析可以满足误差需求。最后,从学生专注占比程度与专注趋势两个维度做出线上课堂学生专注度评价,有助于教师获取实时、精准的教学反馈,从而及时改进教学策略,提高课堂教学质量。

The emotional communication between teachers and students in the online teaching environment has become a vacuum,however,the emotional communication between teachers and students is an important factor affecting the effectiveness of teaching・Usually,"concentration"and"mind—wandering"are two indistinguishable student classroom expressions.The recognition of students'"concentration"and"mind-wandering"similar expressions in the online teaching environment helps teachers to perceive the students,listening status in a timely and accurate manner.In this paper,a method based on the detection of key points in Dlib's eyes is used to identify students5"concentration"and"mind-wandering"expressions,and on this basis,a statistical algorithm for the duration of student expressions is established to calculate the duration of each expression.Experimental results show that the accuracy of the above two expression recognition using key points of the eye is 95.1%,and the data volume of the method is small,the calculation speed is fast,and the result analysis and duration statistical analysis can meet the error requirements.Finally,the evaluation of student concentration in online classrooms from the two dimensions of student concentration and concentration trend helps teachers to obtain real-time and accurate teaching feedback,improve teaching strategies in a timely manner,and improve the quality of classroom teaching.