With the advantages of real-time analysis and visual evaluation results,intelligent technology-enabled teaching behavior evaluation has gradually become a powerful means to help teachers adjust teaching behaviors and ...With the advantages of real-time analysis and visual evaluation results,intelligent technology-enabled teaching behavior evaluation has gradually become a powerful means to help teachers adjust teaching behaviors and improve teaching quality.However,at present,the evaluation of intelligent teachers’behaviors is still in the preliminary exploration stage,and the application research is not deep enough.This paper analyzes the application of intelligent technology in the evaluation of teachers’classroom teaching behaviors from the perspectives of evaluation data,methods,and results.Voice print recognition technology is used to recognize the teachers’identities and track the speech in the classroom videos,and the videos are segmented.Then,the evaluation framework of teachers’classroom teaching behaviors is constructed using three dimensions of emotion,posture,and position preference.Finally,evaluation results are presented to teachers in a more intuitive and easy-to-understand visual way,to help teachers reflect on teaching.This paper aims to promote the transformation of teachers’classroom teaching behavior evaluation toward an intelligent,efficient,and sustainable direction through current research.展开更多
文摘探究教师注意力对于评估课堂教师行为具有极其重要的研究价值。然而,现有的教师注意力识别算法存在无法应对极端头部姿态角度等问题。为此,提出一种基于6DRep Net360模型的教师注意力状态识别算法,提升极端角度中头部姿态估计算法的准确性。相较于传统的依赖条件判断来分类教师注意力状态的方法,设计一种基于支持向量机(SVM)的教师注意力分类模型,对复杂头部姿态角度进行注意力状态的精准识别。为进一步解决算法稳定性和准确性带来的误差数据,提出基于滑动窗口的数据清洗算法,有效提高整体识别结果的真实性和可靠性。通过在构建的CCNUTeacherS tat e数据集上进行一系列的算法评估,实验结果表明,所提出的教师注意力识别算法在CCNUTeacherS tate数据集上达到了90.67%的准确率。
文摘With the advantages of real-time analysis and visual evaluation results,intelligent technology-enabled teaching behavior evaluation has gradually become a powerful means to help teachers adjust teaching behaviors and improve teaching quality.However,at present,the evaluation of intelligent teachers’behaviors is still in the preliminary exploration stage,and the application research is not deep enough.This paper analyzes the application of intelligent technology in the evaluation of teachers’classroom teaching behaviors from the perspectives of evaluation data,methods,and results.Voice print recognition technology is used to recognize the teachers’identities and track the speech in the classroom videos,and the videos are segmented.Then,the evaluation framework of teachers’classroom teaching behaviors is constructed using three dimensions of emotion,posture,and position preference.Finally,evaluation results are presented to teachers in a more intuitive and easy-to-understand visual way,to help teachers reflect on teaching.This paper aims to promote the transformation of teachers’classroom teaching behavior evaluation toward an intelligent,efficient,and sustainable direction through current research.