AI-Powered Driver Drowsiness and Distraction Detection for Enhanced Road Safety

Abstract: Driver drowsiness and distractions are leading causes of accidents, making real-time detection essential for safety. This work employs machine learning and deep learning to monitor drivers through facial and behavioral cues. Real-time video processing analyzes Blink Frequency, Maximum Eye Closure Time, and PERCLOS to detect prolonged eye closure, while Yawning Frequency helps assess fatigue and trigger alerts.Head Pose estimation tracks Euler angles to identify distractions like backseat conversations, and the system detects mobile phone usage without Bluetooth. EAR ensures the driver remains focused. By combining video analysis, image processing, and deep learning, the system enhances road safety, tackling efficiency and accuracy challenges and advancing intelligent transportation.

Keywords: Driver monitoring, drowsiness detection, distraction detection, machine learning, deep learning, image processing, head pose estimation, PERCLOS, Eye Aspect Ratio (EAR), real-time video analysis, road safety.