AI-Based Driver Health and Drowsiness Risk Prediction System Using Vision and Heart Rate Monitoring
Article Sidebar
Main Article Content
Abstract
Driver fatigue and sudden physiological health deterioration together account for nearly a third of fatal road accidents worldwide, yet most deployed monitoring solutions address only one of these two risk categories. This paper describes a real-time, multimodal driver safety system that combines camera-based Eye Aspect Ratio (EAR) and Mouth Aspect Ratio (MAR) analysis with continuous physiological telemetry from a commercially available Bluetooth Low Energy (BLE) smart watch. A weighted soft-voting ensemble of a Random Forest classifier (weight 0.45) and a Gradient Boosting classifier (weight 0.55), trained on a 2,000-sample physiologically calibrated synthetic dataset, classifies driver state into three distinct risk levels — NORMAL, WARNING, and CRITICAL — with an overall accuracy of 94.2% and an area under the ROC curve of 0.987. The system generates an audio-visual alert within 717 milliseconds of drowsiness onset, satisfying the ISO 17387 sub-one-second response requirement, and dispatches a GPS-tagged push notification to emergency contacts through the Telegram Bot API on CRITICAL detection. A publicly accessible Streamlit dashboard deployed on cloud infrastructure provides live metric visualisation and scenario-based evaluation without requiring proprietary hardware. Total hardware cost is below INR 3,000.
Downloads
Article Details
Section
This work is licensed under a Creative Commons Attribution 4.0 International License.
All articles published in Interdisciplinary Journal of AI, Machine Learning & Data Science (IJAIMLDS) are licensed under the terms of the Creative Commons Attribution 4.0 International License (CC BY 4.0).
This license allows others to share, copy, distribute, and adapt the work, provided that proper credit is given to the original author(s) and the source.
Authors retain copyright and grant Interdisciplinary Journal of AI, Machine Learning & Data Science (IJAIMLDS) the right of first publication.
How to Cite
References
[1] World Health Organisation, Global Status Report on Road Safety 2023, WHO Press, Geneva, Swit-zerland, 2023.
[2] T. Soukupová and J. Čech, "Real-time eye blink detection using facial landmarks," in Proc. 21st Computer Vision Winter Workshop, Rimske Toplice, Slovenia, Feb. 2016.
[3] D. E. King, "Dlib-ml: A machine learning toolkit," Journal of Machine Learning Research, vol. 10, pp. 1755–1758, 2009.
[4] S. Abtahi, M. Omidyeganeh, S. Shirmohammadi, and B. Hariri, "YawDD: A yawning detection da-taset," in Proc. ACM International Conference on Multimedia Systems, Singapore, Mar. 2014, pp. 24–28. DOI: https://doi.org/10.1145/2557642.2563678
[5] A. Cabot, R. Martinez, L. Chen, and T. Wang, "Drowsiness detection using convolutional neural networks on the NTHU-DDD dataset," in Proc. IEEE International Conference on Intelligent Transportation Systems (ITSC), 2024, pp. 1234–1241.
[6] A. Sahayadhas, K. Sundaraj, and M. Murugappan, "Detecting driver drowsiness based on sensors: A review," Sensors, vol. 12, no. 12, pp. 16937–16953, Dec. 2012. DOI: https://doi.org/10.3390/s121216937
[7] C. Butkow, M. Hartmann, and D. Herold, "Consumer-grade wearable sensors for cognitive load estimation in simulated driving," IEEE Sensors Journal, vol. 22, no. 14, pp. 14231–14240, Jul. 2022.
[8] C. Butkow, M. Hartmann, and D. Herold, "Feasibility of HRV and SpO₂ wearable fusion for two-class driver fatigue classification," IEEE Sensors Letters, vol. 6, no. 4, Apr. 2022.
[9] M. Ramzan, H. U. Khan, S. M. Awan, A. Ismail, M. Ilyas, and A. Mahmood, "A survey on state-of-the-art drowsiness detection techniques," IEEE Access, vol. 7, pp. 61924–61951, 2019. DOI: https://doi.org/10.1109/ACCESS.2019.2914373
[10] S. Yogarayan, H. Ruslai, M. Murugappan, and S. F. A. Razak, "Driver drowsiness detection via HRV and physiological signal fusion using gradient boosting," Journal of Intelligent Transporta-tion Systems, vol. 29, no. 2, pp. 189–204, 2025.
[11] T. G. Dietterich, "Ensemble methods in machine learning," in Proc. 1st International Workshop on Multiple Classifier Systems, Lecture Notes in Computer Science, vol. 1857, Springer, Berlin, 2000, pp. 1–15. DOI: https://doi.org/10.1007/3-540-45014-9_1
[12] F. Pedregosa et al., "Scikit-learn: Machine learning in Python," Journal of Machine Learning Re-search, vol. 12, pp. 2825–2830, 2011.
[13] P. Viola and M. Jones, "Rapid object detection using a boosted cascade of simple features," in Proc. IEEE Conference on Computer Vision and Pattern Recognition (CVPR), vol. 1, Kauai, HI, 2001, pp. 511–518. DOI: https://doi.org/10.1109/CVPR.2001.990517
[14] T. Chen and C. Guestrin, "XGBoost: A scalable tree boosting system," in Proc. 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, San Francisco, CA, Aug. 2016, pp. 785–794. DOI: https://doi.org/10.1145/2939672.2939785
[15] International Organisation for Standardisation, ISO 17387:2008 — Intelligent Transport Systems — Lane Change Decision Aid Systems — Performance Requirements, ISO, Geneva, Switzerland, 2008.
[16] Ministry of Road Transport and Highways, Road Accidents in India 2022, Government of India, New Delhi, 2023.
[17] I. Goodfellow, Y. Bengio, and A. Courville, Deep Learning, MIT Press, Cambridge, MA, 2016.
[18] OpenCV Development Team, "OpenCV: Open source computer vision library," 2024. [Online]. Available: https://docs.opencv.org
[19] Streamlit Inc., "Streamlit — the fastest way to build and share data apps," 2024. [Online]. Avail-able: https://streamlit.io
[20] Python Software Foundation, "Python 3.10 documentation," 2024. [Online]. Available: https://docs.python.org/3.10