Iot-Enabled Real-Time Driver Monitoring System for Drowsiness and Alcohol Detection Using Arduino and OpenCV

Dept. of CSE (loT & CSBT), East Point College of Engineering and Technology, Bengaluru, Karnataka, India

Published online: 16 April 2026

Abstract

Driver drowsiness and alcohol impairment account for approximately 30-35% of serious road crash incidents globally, claiming over 400,000 lives annually (World Health Organization). This paper presents a real-time, non-intrusive driver impairment detection system that concurrently monitors drowsiness through computer vision and alcohol intoxication through embedded sensing. Drowsiness detection employs the Eye Aspect Ratio (EAR) derived from dlib’s 68-point facial landmark regression, achieving 92.4% accuracy (F1-score = 89.7%, ROC AUC = 0.962) across 1,443 test frames. Alcohol intoxication is continuously monitored via an MQ-3 semiconductor gas sensor calibrated to the 0.05% BAC legal threshold. A three-stage hierarchical response model escalates from auditory alerting through controlled deceleration to autonomous leftlane parking with GSM emergency notification. The complete hardware subsystem is priced at USD 62-78, demonstrating practical deployability for broad vehicular safety applications.