Driver drowsiness is a major cause of road accidents, leading to significant risks for both drivers and passengers. Early detection of fatigue can help prevent such incidents and improve overall road safety. This paper presents a real-time driver drowsiness detection system based on computer vision techniques using MediaPipe FaceMesh and Eye Aspect Ratio (EAR). The proposed system continuously monitors the driver's eye movements and identifies drowsiness by detecting prolonged eye closure. A threshold-based approach is used, where an alert is triggered if the EAR value remains below 0.25 for more than 20 consecutive frames. The system is designed to be lightweight, non-intrusive, and capable of operating in real time on standard hardware without requiring specialized sensors. Experimental results demonstrate that the proposed system achieves an accuracy of approximately 93\% with a low false positive rate, while maintaining an average detection delay of around 150 ms. Compared to deep learning-based approaches, the proposed method offers a computationally efficient and cost-effective solution suitable for real-time applications. The results indicate that the system can effectively detect driver drowsiness and has strong potential for deployment in practical scenarios such as in-vehicle safety systems and fleet monitoring.
Lokanadha Manikanta Kunchala (2026) studied this question.