The performance of students' classroom behavior is an important part of the evaluation of classroom teaching, and the recognition of students' classroom behavior is of great significance to the evaluation of classroom teaching. However, due to the complexity of students' classroom behavior, it is difficult to identify intelligent students' classroom behavior. Therefore, this paper proposes a classroom behavior analysis and evaluation system based on deep learning face recognition technology. The classroom behavior analysis and evaluation system judges whether students pay attention to class from three aspects: Students' side face concentration, students' head up and head down concentration, and eyes opening and closing concentration, so as to provide an objective evaluation basis for students' classroom behavior evaluation in classroom teaching.At present, to solve the problem of performance degradation of convolutional neural network with the deepening of network layers, a deep residual network based on residual structure is proposed. Comparing the accuracy of the depth residual network with that of the depth convolution neural network on this data set, the experimental results show that the former has better network recognition performance.

Wang Ling. Automatic Recognition of Students' Classroom Behavior Based on Computer Vision. Academic Journal of Computing & Information Science (2022), Vol. 5, Issue 2: 31-34. https://doi.org/10.25236/AJCIS.2022.050205.

[1] Xu Jiazhen, Deng Wei, Wei Yantao. Automatic recognition of student classroom behavior based on human skeleton information extraction [J]. Modern Educational Technology, 2020, 30(5):6.

[2] Zhao Long. Recognition and analysis of abnormal behaviors of students in class based on image big data [J]. Information Weekly, 2019(46):1.

[3] Guo Junqi, Lu Jiahao, Wang Ruhan, et al. Recognition of teacher and student classroom behavior driven by deep learning model[J]. Journal of Beijing Normal University (Natural Science Edition), 2021, 57:1-9.

[4] Song Xiaoyu. Teaching practice exploration of experiencing computer vision in the classroom [J]. China Information Technology Education, 2020(21):2.

[5] Zhang Zeping, Yang Lang, Xie Zhixing. Real-time emotion recognition based on Xception in classroom quality analysis [J]. Automation Technology and Application, 2020, 39(6):6.

[6] Xu Jiazhen, Deng Wei, Wei Yantao. Automatic recognition of student classroom behavior based on human skeleton information extraction [J]. 2021(2020-5):108-113.

[7] Xu Man, Zhang Rubi and Francium. Application of management system based on dynamic face recognition in student classroom [J]. Electroacoustic Technology, 2018, 42(9):3.

[8] Liu Qiuhui, Liang Mingxiu, Wang Lin. A Study on the Learning State of Classroom Students Based on Deep Learning [J]. Computer Science and Applications, 2020, 10(12):7.

[9] Zhao Jing, Lu Liqun. Teaching Reform Practice of Computer Vision Inspection and Image Processing Course [J]. University Education, 2017(8):3.

[10] Niu Zuodong, Li Handong. Design of classroom real-time attendance system based on computer vision [J]. Electronic Testing, 2020(4):3.