Real-Time Driver Drowsiness Detection Using Facial Biometrics

Driver drowsiness is a progressive loss of vigilance whose visual manifestations may appear before a critical driving error. This paper presents an interpretable real-time framework that combines Eye Aspect Ratio (EAR), Mouth Aspect Ratio (MAR), Percentage of Eye Closure (PERCLOS), head-pose estimation, and temporal fusion to distinguish Normal, Yawning, Fatigue, and Microsleep states. The NTHU Driver Drowsiness Detection (NTHU-DDD) benchmark is used as the principal experimental reference, following subject-separated training, evaluation, and test partitions to reduce identity leakage. The proposed approach integrates frame-level facial measurements with temporal persistence and multi-cue decision rules, allowing short normal behaviors to be distinguished from sustained drowsiness-related events. Evaluation is performed at the event level over 2,000 driver-state events using accuracy, precision, recall, F1-score, false alarms per hour, and detection latency. The multi-cue framework achieves an overall accuracy of 91.65% and a macro F1-score of 91.41%, with 1.8 false alarms per hour and a mean detection latency of 0.84 s. Ablation analysis further indicates that temporal eye-closure information plays an important role in classification performance, while the complete multi-cue configuration reduces false alarms. The results support the use of interpretable facial and temporal information for real-time driver monitoring, while further validation under natural driving conditions remains necessary.

Driver drowsiness detection; EAR; MAR; PERCLOS; temporal fusion; driver monitoring.

1. Abtahi, S., Omidyeganeh, M., Shirmohammadi, S., & Hariri, B. (2014). YawDD: A yawning detection dataset. Proceedings of the 5th ACM Multimedia Systems Conference (MMSys ’14), 24–28. https://doi.org/10.1145/2557642.2563678 [Google Scholar] [Crossref]

2. Bakheet, S., & Al-Hamadi, A. (2021). A framework for instantaneous driver drowsiness detection based on improved HOG features and Naïve Bayesian classification. Brain Sciences, 11(2), 240. https://doi.org/10.3390/brainsci11020240 [Google Scholar] [Crossref]

3. Bakiasi, V., Kapçiu, R., Zeqo, E., Bushati, S., Bushi, F., Bame, O., & Lamaj, B. (2025). Real-time micro-expression recognition using YOLOv8 and FER2013 dataset. Journal of Information Systems Engineering and Management, 10(18s), 53–64. https://doi.org/10.52783/jisem.v10i18s.2883 [Google Scholar] [Crossref]

4. Centers for Disease Control and Prevention, National Institute for Occupational Safety and Health. (2024). Driver fatigue on the job. https://www.cdc.gov/niosh/motor-vehicle/driver-fatigue/index.html [Google Scholar] [Crossref]

5. Dinges, D. F., Mallis, M. M., Maislin, G., & Powell, J. W. (1998). Evaluation of techniques for ocular measurement as an index of fatigue and as the basis for alertness management (Report No. DOT-HS-808-762). National Highway Traffic Safety Administration. https://rosap.ntl.bts.gov/view/dot/2518 [Google Scholar] [Crossref]

6. Fonseca, T., & Ferreira, S. (2025). Drowsiness detection in drivers: A systematic review of deep learning-based models. Applied Sciences, 15(16), 9018. https://doi.org/10.3390/app15169018 [Google Scholar] [Crossref]

7. Ghoddoosian, R., Galib, M., & Athitsos, V. (2019). A realistic dataset and baseline temporal model for early drowsiness detection. Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops. https://doi.org/10.1109/CVPRW.2019.00027 [Google Scholar] [Crossref]

8. Google. (2020). MediaPipe Face Mesh and Face Geometry. https://github.com/google-ai-edge/mediapipe/blob/master/docs/solutions/face_mesh.md [Google Scholar] [Crossref]

9. INSTAT. (2026). Road accidents, December 2025. Institute of Statistics, Albania. [Google Scholar] [Crossref]

10. https://www.instat.gov.al/en/themes/industry-trade-and-services/transport-accidents-and-characteristics-of-road-vehicles/publikimet/2025/road-accidents-december-2025/ [Google Scholar] [Crossref]

11. Moredo, M. J. R., Celino, J. D. S., & Ibarra, J. B. G. (2025). Multi-feature long short-term memory facial recognition for real-time automated drowsiness observation of automobile drivers with Raspberry Pi 4. Engineering Proceedings, 92(1), 52. https://doi.org/10.3390/engproc2025092052 [Google Scholar] [Crossref]

12. National Highway Traffic Safety Administration. (2026). Drowsy driving: Avoid falling asleep behind the wheel. https://www.nhtsa.gov/risky-driving/drowsy-driving [Google Scholar] [Crossref]

13. Peker, N. Y., Peker, E., & Zengin, A. (2026). A comparative review of driver drowsiness detection systems: Recent advances, applications, challenges, and future directions. Measurement, 279, 121676. [Google Scholar] [Crossref]

14. https://doi.org/10.1016/j.measurement.2026.121676 [Google Scholar] [Crossref]

15. Shtino, V. B., Muça, M., & Bushati, S. (2024). Machine learning integration for precise facial micro-expression recognition. Journal of Computer Science, 20(11), 1545–1558. [Google Scholar] [Crossref]

16. https://doi.org/10.3844/jcssp.2024.1545.1558 [Google Scholar] [Crossref]

17. Shtino, V.B. (2026). Enhancing facial expression and micro-expression recognition through hybrid dimensionality reduction and machine learning classification. International Journal of Research and Innovation in Applied Science, 11(7), 2066–2082. https://doi.org/10.51584/IJRIAS.2026.11070148 [Google Scholar] [Crossref]

18. Soukupová, T., & Čech, J. (2016). Real-time eye blink detection using facial landmarks. Proceedings of the 21st Computer Vision Winter Workshop, 1–8. [Google Scholar] [Crossref]

19. https://cmp.felk.cvut.cz/~cechj/publications/CVWW2016.pdf [Google Scholar] [Crossref]

20. Weng, C.-H., Lai, Y.-H., & Lai, S.-H. (2017). Driver drowsiness detection via a hierarchical temporal deep belief network. In Computer Vision – ACCV 2016 Workshops (LNCS 10118, pp. 117–133). Springer. https://doi.org/10.1007/978-3-319-54526-4_9 [Google Scholar] [Crossref]

21. Zeqo, E., (Shtino), V. B., (Lamaj), B. M., & Dyrkaj, F. (2026). Recent advances and challenges in facial emotion recognition systems: A review of deep learning approaches. International Journal for Research Trends and Innovation, 11(5), a620–a631. https://doi.org/10.56975/ijrti.v11i5.212555 [Google Scholar] [Crossref]

22. Zhao, G., He, Y., Yang, H., & Tao, Y. (2022). Research on fatigue detection based on visual features. IET Image Processing, 16(4), 1044–1053. https://doi.org/10.1049/ipr2.12207 [Google Scholar] [Crossref]

• Enhanced Multi-Task CNN For Age, Gender, Race with Mask in Facial Images