Intelligent Driver Drowsiness Vigilance Monitoring Using Machine Learning

Abstract

Driver drowsiness is a significant contributing factor to road accidents, posing risks to both life and property. This research presents a novel approach to mitigate these risks through the development of a machine learning-based drowsiness detection system. Leveraging datasets such as the MRL Eye dataset, our system employs real-time monitoring of driver eye movements and facial expressions to identify early signs of drowsiness. Importantly, we incorporate temporal dynamics analysis, focusing on subtle variations in pupil diameter, gaze position, and eyelid conditions over time, to enhance the system's accuracy in predicting drowsiness. Upon detection of drowsiness indicators, the system triggers proactive alerts in the form of sound and vibration, prompting the driver to take corrective action.