Student Concentration Recognition Model in English Classroom Based on ResNet50 andVGG16

Authors

  • Dongdong Tang Qingdao Huanghai University, Qingdao 266427, China
  • Qingqing Jiang Haidu College, Qingdao Agricultural University, Yantai 265200, China

Abstract

Students’ classroom behavior can indicate their listening status in real time, and teachers judge students’ listening status based on their own experience and students’ behavior. However, this method greatly tests teachers’ teaching experience and energy, and it is difficult to achieve good results on this basis. Therefore, this study proposes an English classroom student focus recognition model based on deep residual networks and visual geometry group networks. The model selects the “You Only Look OnceV3” algorithm as the target position detector, and then uses deep residual networks and visual geometry group networks to recognize and classify student classroom behavior, thereby determining each student’s class state. The experimental
results show that the accuracy of the cropped model is significantly higher than that before cropping. When the number of iterations reaches 500, the accuracy of the model before and after image cropping in the visual geometry group network model is 0.88 and 0.97, respectively. In regard to the deep residual network model, the accuracy of the model before and after image cropping is 0.86 and 0.98, respectively. For the dual network hybrid model, the model accuracy before and after image cropping was 0.90 and 0.99, respectively. The research results indicate that the proposed dual network hybrid algorithm model has excellent performance in recognizing student state.

Keywords: Behavior recognition, transfer learning, ResNet50,VGG16, concentration

Cite As

D. Tang, Q. Jiang, "Student Concentration Recognition Model in English Classroom Based on ResNet50 andVGG16",
Engineering Intelligent Systems, vol. 33 no. 2, pp. 201-211, 2025.



Downloads

Published

2025-03-01

Issue

Vol. 33 No. 2 (2025): International Journal of Engineering Intelligent Systems

Section

Article

License

The submission of a paper implies that, if accepted for publication, it will not be published elsewhere in the same form, in any language, without the prior consent of the publisher. Before publication, authors are requested to assign copyright to CRL Publishing Ltd. This allows CRL to sanction photocopying, and to authorize the reprinting of issues or volumes according to demand. Authors' traditional rights will not be jeopardized by assigning Copyright in this way, as they retain the right to reuse the material following publication, and to veto third-party publication.