Abstract :

Fatigue-induced And Distraction-driven Road Accidents Continue To Claim Thousands Of Lives Globally, Motivating The Development Of Intelligent, Camera-based Driver Monitoring Frameworks. This Paper Presents A Real-time Driver Drowsiness Detection And Alert System That Fuses Three Complementary Analysis Modules: A Gaze Score Module, A Pose Estimation Module, And A Deep Learning-based Drowsiness Detection Module. The Gaze Score Component Continuously Evaluates Ocular Fixation And Line-of-sight Deviation To Assess Road-focused Attention. The Pose Estimation Component Computes Euler Angles—pitch, Yaw, And Roll—to Identify Head-orientation Anomalies Indicative Of Distraction Or Inattentiveness. The Drowsiness Detection Component Leverages A Shallow Convolutional Neural Network To Identify Micro-sleep Episodes, Protracted Eyelid Closure, And Fatigue-related Blink Dynamics With Resilience Across Variable Illumination And Occlusion Scenarios. Eye Aspect Ratio (EAR) And PERCLOS Metrics Serve As Quantitative Indicators Of Drowsiness Severity. Individual Module Outputs Are Merged By A Decision Unit That Dynamically Computes A Composite Risk Score And Triggers Context-sensitive Audio-visual Alerts Before Impairment Reaches A Critical Threshold. Experimental Evaluation Confirms Stable Frame-rate Performance, Low-latency Alert Generation, And Robust Detection Under Diverse Real-world Driving Conditions. The Integrated Approach Advances Beyond Single-cue Detection Paradigms, Delivering A Practical, Adaptive, And Hardware-efficient Solution That Meaningfully Contributes To Safer, More Intelligent Transportation Systems. Index Terms—Driver Monitoring, Drowsiness Detection, Deep Learning, Eye Aspect Ratio, Gaze Score, Pose Estimation, Real-Time Video Analysis, Road Safety.

Published:

28-3-2026

Issue:

Vol. 26 No. 3 (2026)

Page Nos:

894-902

Section:

Articles

License:

This work is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License.

How to Cite