A Novel Approach to Advancing Drowsiness Detection Using Head-Angle and Eye-Aspect Ratio

Drowsiness is a significant risk factor for accidents, particularly in situations requiring sustained attention, such as driving or operating machinery. This research proposes an innovative methodology for detecting driver drowsiness using advanced computer vision algorithms and techniques. Our system employs a shapeshift predictor and face-tracking algorithms to monitor the driver's facial features and head movements. Specifically, it detects three distinct indicators of drowsiness: excessive head tilting to the right or left, significant drooping of the chin, and prolonged eye closure (defined as eyes remaining closed for more than 24 consecutive frames). When any of these conditions are met, an alarm is triggered, alerting the driver to their potentially hazardous state. The proposed system offers a robust and non-intrusive solution for real-time drowsiness monitoring, with potential applications in various fields where vigilance is critical for safety.