Deep Learning-Based Method for Detecting Cheating Behaviors in Remote Examination Settings

DOI: https://doi.org/10.62517/jbdc.202501205

Author(s)

Rongcong Chen, Yang Chen, Zuozheng Lian*, Guangda Zhang

Affiliation(s)

College of Computer and Control Engineering, Qiqihar University, Qiqihar, Heilongjiang, China *Corresponding Author

Abstract

Examinations serve as a crucial means of assessing students' abilities and knowledge levels, necessitating supervision during the process to ensure fairness. However, most existing examinations rely on multiple invigilators stationed in the examination hall, supplemented by video surveillance and computer equipment, which demands substantial human resources and time costs while exhibiting low invigilation efficiency. To address this issue, this paper proposes a method for detecting cheating behaviors in examination halls, which primarily consists of two modules: face-recognition and Alpha Pose. The former module implements facial recognition to prevent impersonation, while the latter captures key point data and employs the ST-GCNs model to detect examinees' actions, thereby preventing cheating behaviors. This method can be deployed in examination halls equipped with video surveillance and computer devices, demonstrating high recognition rates for common cheating behaviors and effectively reducing the human resources and time costs associated with examinations.

Keywords

Image Processing; Facial Recognition and Detection; Behavior Recognition; Cheating Action Detection

References

[1] Huang et al. Fall detection model based on AlphaPose combined with LSTM and Lightgbm. Proceedings of SPIE, 2024, 13086(000):8. DOI:10.1117/12.2692742. [2] Liu et al. A method for detecting cheating in examination rooms based on line of sight estimation. Journal of Taiyuan Normal University (Natural Science Edition), 2022, 21 (03): 47-52+96 [3] Li. Abnormal behavior detection in examination rooms based on head motion analysis. Changchun University of Science and Technology, 2018. [4] Lin et al. A method for detecting cheating behavior in examination rooms based on background subtraction. Journal of Jinan University (Natural Science Edition), 2015, 29 (06): 406-409. DOI: 0.13349 /j.cnki. jdxbn. 2015.06.002. [5] Yan et al. Spatial Temporal Graph Convolutional Networks for Skeleton-Based Action Recognition. AAAI Conference on Artifi cial Intelligence (2018). [6] N. Dalal & B. Triggs. Histograms of oriented gradients for human detection //2005 IEEE Computer Society Conference on Computer Vision and Pattern Recognition (CVPR'05), San Diego, CA, USA, 2005, pp. 886-893 vol. 1, DOI: 10.1109/CVPR.2005.177. [7] Li et al. A review of human joint motion detection and tracking. Mechanical, 2020,47 (08): 18-23. [8] Singh S et al. Benchmarking Object Detectors with COCO: A New Path Forward // European Conference on Computer Vision. Springer, Cham, 2025.DOI:10.1007 /978-3-031-72784-16. [9] Wei & Gao. Research on Video Surveillance Moving Target Detec tion Algorithm. Electronic Testing, 2016, (07): 53-54. [10] Sheng & Wang. Research on Human Pose Recognition Based on Computer Vision. Modern Information Technology, 2022,6 (16): 87-91+95. DOI: 10.19850/j.cnki. 2096-4706.2022.16.023.