Classroom Anomalous Behavior Detection based on Improved YOLOv5

Fu  Yao

doi:10.6180/jase.202507_28(7).0008

Classroom Anomalous Behavior Detection based on Improved YOLOv5

Computer Science and Information Engineering

Fu YaoThis email address is being protected from spambots. You need JavaScript enabled to view it.

School of Information Science and Engineering, Shenyang Ligong University, Shenyang 110159, China

Received: August 7, 2024
Accepted: September 1, 2024
Publication Date: October 7, 2024

Copyright The Author(s). This is an open access article distributed under the terms of the Creative Commons Attribution License (CC BY 4.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are cited.

Download Citation: ||https://doi.org/10.6180/jase.202507_28(7).0008

In recent years, there has been an increasing interest in using machine learning methods to improve deep learning-based classroom abnormal behaviour detection tasks, and the theory of fractional order calculus is beginning to be used to enhance the model’s ability to describe the features of the data. In this paper, we propose a classroom abnormal behaviour detection method based on fractional order calculus for YOLOv5 to monitor and analyse students’ classroom behaviour immediately. A fractional order coordinate attention mechanism is designed in the YOLOv5 feature learning stage to capture the long-range relationship of features in combination with position information, while a hybrid convolutional layer is introduced to achieve computational lightness, and finally an improved loss function is used for training to improve the detection accuracy robustness. In this paper, we use the improved YOLOv5 model based on fractional order differentiation for classroom abnormal behaviour detection. It is experimentally shown that the method proposed in this paper achieves significant performance improvement in classroom behaviour detection, with 5.4 Spf improvement in detection speed and 4.1% improvement in classroom abnormal behaviour accuracy, which is highly observable and provides more targeted intervention and management tools for the education industry.

Keywords: YOLOv5; Fractional order differentiation; Deep learning; Abnormal behaviour detection

[1] L. Shi and X. Di, (2023) “A recognition method of learning behaviour in English online classroom based on feature data mining" International Journal of Reasoningbased Intelligent Systems 15: 8–14. DOI: 10.1504/IJRIS.2023.128375.
[2] T. Guo, X. Bai, X. Tian, S. Firmin, and F. Xia, (2022) “Educational anomaly analytics: features, methods, and challenges" Frontiers in big Data 4: 811840. DOI: 10.3389/fdata.2021.811840.
[3] Y. Liu, H. Chen, and A. Thoff, (2020) “Research on evaluation method of students’ classroom performance based on artificial intelligence" International Journal of Continuing Engineering Education and Life Long Learning 30: 476–491. DOI: 10.1504/IJCEELL.2020. 110925.
[4] Y. Xie, S. Zhang, and Y. Liu, (2021) “Abnormal Behavior Recognition in Classroom Pose Estimation of College Students Based on Spatiotemporal Representation Learning" Traitement du Signal 38: 89–95. DOI: 10.18280/ ts.380109.
[5] X. Zhang, (2022) “A Gaussian High-Dimensional Random Matrix-Based Method for Detecting Abnormal Student Behaviour in Chinese Language Classrooms" Mathematical Problems in Engineering 2022: 6957097. DOI: 10.1155/2022/6957097.
[6] S. Zhang, H. Liu, C. Sun, X. Wu, P. Wen, F. Yu, and J. Zhang, (2023) “MSTA-SlowFast: A student behavior detector for classroom environments" Sensors 23: 5205. DOI: 10.3390/s23115205.
[7] M. A. E. Abbas and S. Hameed, (2022) “A Systematic Review of Deep Learning Based Online Exam Proctoring Systems for Abnormal Student Behaviour Detection" International Journal of Scientific Research in Science, Engineering and Technology 9: 192. DOI: 10.32628/IJSRSET229428.
[8] C. Pabba and P. Kumar, (2022) “An intelligent system for monitoring students’ engagement in large classroom teaching through facial expression recognition" Expert Systems 39: e12839. DOI: 10.1111/exsy.12839.
[9] J. Zhang, Z. Zhang, L. Guan, and H. Hu. Research on Classroom Behavior Recognition and Detection Method Based on Deep Learning. 2024. DOI: 10.1109/cvidl62147.2024.10604092.
[10] J. Redmon, S. Divvala, R. Girshick, and A. Farhadi. “You only look once: Unified, real-time object detection”. In: Proceedings of the IEEE conference on computer vision and pattern recognition. 2015, 779–788. DOI: 10.48550/arXiv.1506.02640.
[11] J. Redmon and A. Farhadi. “YOLO9000: better, faster, stronger”. In: Proceedings of the IEEE conference on computer vision and pattern recognition. 2017, 7263–7271. DOI: 10.1109/CVPR.2017.690.
[12] L. Tang, T. Xie, Y. Yang, and H. Wang, (2022) “Classroom behavior detection based on improved YOLOv5 algorithm combining multi-scale feature fusion and attention mechanism" Applied Sciences 12: 6790. DOI: 10.3390/app12136790.
[13] J. Wen, Y. Qin, and S. Hu. “Abnormal behavior identification of examinees based on improved YOLOv5”. In: International Conference on Computer Graphics, Artificial Intelligence, and Data Processing (ICCAID 2022). 12604. 2022, 946–953. DOI: 10.1117/12.2674630.
[14] Z. Zhang, D. Ao, L. Zhou, X. Yuan, and M. Luo. “Laboratory behavior detection method based on improved Yolov5 model”. In: 2021 International Conference on Cyber-Physical Social Intelligence (ICCSI). 2021, 1–6. DOI: 10.1109/ICCSI53130.2021.9736251.
[15] F. Lei, F. Tang, and S. Li, (2022) “Underwater target detection algorithm based on improved YOLOv5" Journal of Marine Science and Engineering 10: 310. DOI: 10.3390/jmse10030310.
[16] R. Girshick, J. Donahue, T. Darrell, and J. Malik. “Rich feature hierarchies for accurate object detection and semantic segmentation”. In: Proceedings of the IEEE conference on computer vision and pattern recognition. 2014, 580–587. DOI: 10.1109/CVPR.2014.81.
[17] Q. Zheng, M. Yang, X. Tian, X. Wang, and D. Wang, (2020) “Rethinking the Role of Activation Functions in Deep Convolutional Neural Networks for Image Classification" engineering letters 28: 80.
[18] J. Bai, J. Dai, Z. Wang, and S. Yang, (2022) “A detection method of the rescue targets in the marine casualty based on improved YOLOv5s" Frontiers in Neurorobotics 16: DOI: 10.3389/fnbot.2022.1053124.
[19] C. Chen, F. Wang, Y. Cai, S. Yi, and B. Zhang, (2023) “An improved YOLOv5s-based Agaricus bisporus detection algorithm" Agronomy 13: 1871. DOI: 10.3390/ agronomy13071871.
[20] M. Gong, D. Wang, X. Zhao, H. Guo, D. Luo, and M. Song. “A review of non-maximum suppression algorithms for deep learning target detection”. In: Seventh Symposium on Novel Photoelectronic Detection Technology and Applications. 11763. SPIE, 2021, 821–828.
[21] H. Li, J. Li, H. Wei, Z. Liu, Z. Zhan, and Q. Ren, (2022) “Slim-neck by GSConv: A better design paradigm of detector architectures for autonomous vehicles" arXiv: DOI: 10.48550/arXiv.2206.02424.
[22] Q. Hou, D. Zhou, and J. Feng. “Coordinate attention for efficient mobile network design”. In: Proceedings of the IEEE/CVF conference on computer vision and pattern recognition. 2021, 13713–13722. DOI: 10.1109/CVPR46437.2021.01350.
[23] Z. Zheng, P. Wang, W. Liu, J. Li, R. Ye, and D. Ren. “Distance-IoU loss: Faster and better learning for bounding box regression”. In: Proceedings of the AAAI conference on artificial intelligence. 34. 07. 2020, 12993–13000. DOI: 10.1609/aaai.v34i07.6999.
[24] Y.-F. Zhang, W. Ren, Z. Zhang, Z. Jia, L. Wang, and T. Tan, (2022) “Focal and efficient IOU loss for accurate bounding box regression" Neurocomputing 506: 146– 157. DOI: 10.48550/arXiv.2101.08158.
[25] Z. Y. Khan and Z. Niu, (2021) “CNN with depthwise separable convolutions and combined kernels for rating prediction" Expert Systems with Applications 170: 114528. DOI: 10.1016/j.eswa.2020.114528.
[26] H. Srivastava and K. Sarawadekar. “A depthwise separable convolution architecture for CNN accelerator”. In: 2020 IEEE Applied Signal Processing Conference (ASPCON). 2020, 1–5. DOI: 10.1109/ASPCON49795.2020.9276672.
[27] R. Girshick. “Fast r-cnn”. In: Proceedings of the IEEE international conference on computer vision. 2015, 1440– 1448. DOI: 10.1109/ICCV.2015.1691.
[28] S. Ren, K. He, R. Girshick, and J. Sun, (2016) “Faster R-CNN: Towards real-time object detection with region proposal networks" IEEE transactions on pattern analysis and machine intelligence 39: 1137–1149. DOI: 10.1109/TPAMI.2016.2577031.
[29] J. Redmon and A. Farhadi, (2018) “YOLOv3: An Incremental Improvement" ArXiv abs/1804.02767: DOI: 10.48550/arXiv.1804.02767.
[30] W. Liu, D. Anguelov, D. Erhan, C. Szegedy, S. Reed, C.-Y. Fu, and A. C. Berg. “Ssd: Single shot multibox detector”. In: Computer Vision–ECCV 2016: 14th European Conference, Amsterdam, The Netherlands, October 11–14, 2016, Proceedings, Part I 14. 2016, 21–37. DOI: 10.48550/arXiv.1512.02325.