Author(s)

Yao Tian, Cheng Peng^*

Affiliation(s)

School of Computer Science and Technology, Xinjiang Normal University, Urumqi 830054, Xinjiang, China.

Corresponding Author

Cheng Peng

ABSTRACT

Traditional methods of teaching behavior recording suffer from inefficiency, long data mining times, and large computational workloads for statistical analysis. Large models and artificial intelligence offer new technical solutions that can significantly improve teaching quality and optimize the teaching process. This study, based on the improved YOLOv11 model, presents a fine-grained teaching behavior recognition technology aimed at addressing the challenges in smart classroom environments. In response to the complexity of classroom environments and the high similarity of teaching behaviors, an improved YOLOv11 algorithm is proposed. The algorithm introduces the MSCB (Multi-Scale Context Block) and SCSA (Spatial-Channel Self-Attention) modules to enhance the robustness of the model's recognition capabilities. Experimental results show that the improved model performs better in teacher behavior detection, with higher accuracy and efficiency, offering a new approach to teaching behavior recognition.

KEYWORDS

Teaching behaviors; Object detection; Fine-grained recognition; YOLOv11; Intelligent teaching

CITE THIS PAPER

Yao Tian, Cheng Peng. Enhancing the YOLOv11 model for teaching behavior recognition. Journal of Computer Science and Electrical Engineering. 2025, 7(4): 60-65. DOI: https://doi.org/10.61784/jcsee3070.

REFERENCES

[1] Xu Lan, Deng Yingfeng. Research on the Path of Empowering High-Quality Development of Vocational Education Through the “Three Education” Reform—Based on the Background of Industrial Digital Transformation. Vocational Education Forum, 2022(7).

[2] Ren Jiemin. Positioning of Teachers' Roles in Multimedia Teaching Environment Under Constructivist Learning Theory. Curriculum Education Research, 2018(33): 193–194.

[3] Zhang Zhe, Chen Xiaohui, Qin Pengxi, et al. Meta-analysis of Factors Influencing Teachers' Use of Intelligent Technology in Teaching. Modern Distance Education, 2019(2).

[4] Zhao Gang, Zhu Wenjuan, Hu Biling, et al. A simple teacher behavior recognition method for massive teaching videos based on teacher set. Applied Intelligence, 2021, 51(12).

[5] Guo J, Lü J, Wang R, et al. Deep learning model-driven teacher-student classroom behavior recognition. Journal of Beijing Normal University (Natural Science Edition), 2021, 57(06): 905–912.

[6] Ding N. Intelligent analysis and recognition of teacher body movements in secondary school classroom videos. Master's Thesis, Central China Normal University, 2020.

[7] Wang T. Research on classroom teaching behavior analysis methods based on human motion detection. Master's Thesis, Chang'an University, 2020.

[8] Zheng Y. A posture recognition-based method for teacher teaching behavior evaluation. Software Engineering, 2021, 24(04): 6–9.

[9] P Shiyan, Z Anran, L Shuhui, Z Zhiqi. Automatic recognition of teachers' nonverbal behavior based on dilated convolution. 2022 IEEE 5th International Conference on Information Systems and Computer Aided Education (ICISCAE), 2022: 162–167.

[10] Ma X. Research and application of teacher behavior recognition for smart classrooms. Master's Thesis, Yunnan Normal University, 2023.

[11] Liu Y. Research on the evaluation of teaching effectiveness for primary and secondary school teachers in digital education environments. Master's Thesis, Northwest Normal University, 2023.

[12] Carreira J, Zisserman A. Quo vadis, action recognition? A new model and the Kinetics dataset. Computer Vision and Pattern Recognition (CVPR), IEEE Conference on, 2017: 4724–4733.

[13] Kuehne H, Jhuang H, Garrote E, et al. HMDB: A large video database for human motion recognition. In Proc. ICCV, 2011: 2556–2563.

[14] Soomro K, Zamir A, Shah M. UCF101: A dataset of 101 human actions classes from videos in the wild. Computer Science, 2012.

[15] Pang S, Hao J, Hu H, et al. Teacher behavior recognition method based on spatiotemporal graph convolutional neural networks. Journal of Central China Normal University (Natural Science Edition), 2023, 57(05): 715–723.

[16] Liu Q, He H, Wu L, et al. Classroom teaching behavior analysis method based on artificial intelligence and its application. China Educational Technology, 2019(09): 13–21.

[17] Fu D, Zhang H, et al. Educational information processing. Beijing: Beijing Normal University Press, 2011.

[18] Mu S, Zuo P. Research on classroom teaching behavior analysis methods in an information-based teaching environment. Educational Technology Research, 2015, 36(09): 62–69.

[19] Yang F, Wang T. SCB-dataset: A dataset for detecting student classroom behavior, 2023.

[20] Jegham N, Chan Y, Marwan A, et al. Evaluating the evolution of YOLO (You Only Look Once) models: A comprehensive benchmark study of YOLO11 and its predecessors. arXiv preprint arXiv:2411.00201, 2024.

[21] Rahman M, Marculescu R. Medical image segmentation via cascaded attention decoding. IEEE/CVF Winter Conference on Applications of Computer Vision, 2023: 6222–6231.

[22] Sandler M, Howard A, Zhu M, Zhmoginov A, Chen L. Mobilenetv2: Inverted residuals and linear bottlenecks. IEEE Conference on Computer Vision and Pattern Recognition, 2018: 4510–4520.

[23] Zhang X, Zhou X, Lin M, Sun J. Shufflenet: An extremely efficient convolutional neural network for mobile devices. IEEE Conference on Computer Vision and Pattern Recognition, 2018: 6848–6856.

[24] Krizhevsky A, Hinton G. Convolutional deep belief networks on CIFAR-10. Unpublished manuscript, 2010, 40(7): 1–9.

[25] Vaswani A, Shazeer N, Parmar N, et al. Attention is all you need. Advances in Neural Information Processing Systems, 2017, 30.

[26] Ma N, Zhang X, Zheng H, Sun J. Shufflenet v2: Practical guidelines for efficient CNN architecture design. Proceedings of the European Conference on Computer Vision (ECCV), 2018: 116–131.

[27] Han K, Wang Y, Tian Q, et al. Ghostnet: More features from cheap operations. Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, 2020: 1580–1589.

[28] Chen J, Kao S, He H, et al. Run, don’t walk: Chasing higher FLOPs for faster neural networks. Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, 2023: 12021–12031.

[29] Hu J, Shen L, Sun G. Squeeze-and-excitation networks. Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, 2018: 7132–7141.

[30] Hou Q, Zhou D, Feng J. Coordinate attention for efficient mobile network design. Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, 2021: 13713–13722.