Safe Driving Assistant Using Head Pose and Eye Gaze Detection

Abstract—Road accidents caused by driver fatigue anddistraction remain a major concern in transportationsafety worldwide. Prolonged driving hours, lack of rest,and inattentive behaviour significantly impair a driver’salertness, reaction time, and decision-making ability, oftenleading to severe accidents. Traditional vehicle safetymechanisms primarily focus on minimizing post-accidentdamage rather than preventing accidents caused by humanfactors. Therefore, continuous monitoring of driverbehaviour and early detection of unsafe conditions areessential.This paper presents a Safe Driving Assistant system thatdetects driver drowsiness and distraction using Eye AspectRatio (EAR) and head pose estimation techniques based onreal-time facial landmark analysis. The system usesOpenCV and Media pipe to extract facial features from alive camera feed. Drowsiness is identified by analyzingprolonged eye closure, while distraction is detected throughabnormal head orientation and gaze direction. Upondetecting unsafe behavior, the system generates visual,audio, and vibration alerts and communicates with amobile application to send alert messages along with thedriver’s live location to emergency contacts. Experimentalresults show that the proposed system effectively detectsearly signs of fatigue and distraction, thereby enhancingroad safety and supporting intelligent transportationsystems.