Automated recognition of abnormal head behaviors is a practical component of computer-vision-based examination monitoring. This study proposes a lightweight method in which pose geometry provides the primary discriminative information and contour or edge morphology serves as auxiliary evidence. YuNet automatically localizes the facial region, from which 20 contour and edge descriptors and 13 five-point pose-geometric features are extracted. Logistic regression, support vector machine, random forest, and K-nearest neighbors are evaluated using three-fold leave-one-participant-out validation with group-aware inner splitting to reduce identity and adjacent-frame leakage. Experiments use 12,553 images from three classes in the Mendeley Cheating Scenario Dataset in Online Exam. Random forest achieves the highest observed performance, with an accuracy of 0.9766 ± 0.0170 and a macro-F1 score of 0.9653 ± 0.0322. Pose Only reaches 0.9636 ± 0.0238 macro-F1, Contour Only reaches 0.2545 ± 0.0779, and the combined feature set reaches 0.9653 ± 0.0322. The small and participant-dependent gain of the combined representation indicates that contour information is auxiliary rather than universally beneficial. The conclusions are limited to three participants and require validation on larger independent cohorts and real examination environments.

Tianzhen Chen. Recognition of Abnormal Head Behaviors in Online Examinations Using Pose Geometry and Contour-Assisted Features. Academic Journal of Computing & Information Science (2026), Vol. 9, Issue 8: 13-20. https://doi.org/10.25236/AJCIS.2026.090802.

[1] Nigam A, Pasricha R, Singh T, et al. A systematic review on AI-based proctoring systems: Past, present and future[J]. Education and Information Technologies, 2021, 26(5): 6421-6445.

[2] Garg M, Goel A. A systematic literature review on online assessment security: Current challenges and integrity strategies[J]. Computers & Security, 2022, 113: 102544.

[3] Han S, Nikou S, Ayele W Y. Digital proctoring in higher education: A systematic literature review[J]. International Journal of Educational Management, 2023, 38(1): 265-285.

[4] Kaddoura S, Gumaei A. Towards effective and efficient online exam systems using deep learning-based cheating detection approach[J]. Intelligent Systems with Applications, 2022, 16: 200153.

[5] Tejaswi P, Venkatramaphanikumar S, Kishore K V K. Proctor Net: An AI framework for suspicious activity detection in online proctored examinations[J]. Measurement, 2023, 206: 112266.

[6] Verma P, Malhotra N, Suri R, et al. Automated smart artificial intelligence-based proctoring system using deep learning[J]. Soft Computing, 2024, 28(4): 3479-3489.

[7] Hu Z, Jing Y, Wu G, et al. Multi-Perspective Adaptive Paperless Examination Cheating Detection System Based on Image Recognition[J]. Applied Sciences, 2024, 14(10): 4048.

[8] Singh T, Nair R R, Babu T, et al. Enhancing Academic Integrity in Online Assessments: Introducing an Effective Online Exam Proctoring Model using YOLO[J]. Procedia Computer Science, 2024, 235: 1399-1408.

[9] Malhotra M, Chhabra I. MANIT: A multilayer ANN integrated framework using biometrics and historical features for online examination proctoring[J]. Scientific Reports, 2025, 15: 29302.

[10] Kapoor S, Narayanan A. Leakage and the reproducibility crisis in machine-learning-based science[J]. Patterns, 2023, 4(9): 100804.

[11] Wu W, Peng H, Yu S. YuNet: A tiny millisecond-level face detector[J]. Machine Intelligence Research, 2023, 20: 656-665.

[12] Feng Y, Yu S, Peng H, et al. Detect Faces Efficiently: A Survey and Evaluations[J]. IEEE Transactions on Biometrics, Behavior, and Identity Science, 2022, 4(1): 1-18.

[13] Guo J, Deng J, Lattas A, et al. Sample and Computation Redistribution for Efficient Face Detection[C]//International Conference on Learning Representations. 2022.

[14] Liu Y, Wang F, Deng J, et al. MogFace: Towards a Deeper Appreciation on Face Detection[C]//Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition. 2022: 4093-4102.

[15] Li H, Guo Z, Rhee S M, et al. Towards Accurate Facial Landmark Detection via Cascaded Transformers[C]//Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition. 2022: 4176-4185.

[16] Dhingra N. LwPosr: Lightweight Efficient Fine-Grained Head Pose Estimation[C]//Proceedings of the IEEE/CVF Winter Conference on Applications of Computer Vision. 2022: 1204-1214.

[17] Zhang C, Liu H, Deng Y, et al. TokenHPE: Learning Orientation Tokens for Efficient Head Pose Estimation via Transformers[C]//Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition. 2023: 8897-8906.

[18] Canny J. A computational approach to edge detection[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 1986, PAMI-8(6): 679-698.

[19] Hu M K. Visual pattern recognition by moment invariants[J]. IRE Transactions on Information Theory, 1962, 8(2): 179-187.

[20] Rabiha S G. Cheating Scenario Dataset in Online Exam[DS/OL]. Mendeley Data, 2024, Version 1. DOI: 10.17632/mjrfmvsh7d.1.