IoT-Based Driver Safety System with Real-Time Drowsiness and Alcohol Monitoring via Vision and Multi-Sensor Integration

Road traffic accidents, often due to drunk or drowsy driving, cause significant losses. This paper presents an IoTenabled Driver Safety Alert System for real-time alcohol and drowsiness monitoring using an MQ3 alcohol sensor, IR sensor for eye movement detection, and Raspberry Pi with Pi Camera Module for vision-based drowsiness detection via Eye Aspect Ratio (EAR). A dual-mode Arduino Uno and Raspberry Pi architecture enhances modularity and reliability. Tests with 20 participants achieved 97.5 % accuracy, including eyewear cases. Impairment triggers staged alerts alarms, vibrations, lights-and slows/stops the vehicle within $\mathbf{1 0 - 1 5}$ seconds. Vision-based detection sends secure email alerts to authorities, while Bluetooth provides real-time mobile notifications. This low-cost, multimodal system is scalable for intelligent transportation.