DRIVER DROWSINESS DETECTION SYSTEM

Abstract

The Driver Drowsiness Detection (DDD) System is a mobile-based application designed to enhance road safety by detecting driver fatigue in real-time. Addressing the global issue of fatigue-related road accidents, the system leverages deep learning models to analyze facial and physiological indicators, including prolonged eye closure, yawning, and head tilting. It integrates three models: a Baseline Temporal Model (BTM) for blink sequence analysis, an Electrocardiogram (ECG) model for heart rate monitoring, and a new model using Eye Aspect Ratio (EAR), Mouth Aspect Ratio (MAR), and head pose estimation. Built with React Native for the front-end and Django REST Framework with PostgreSQL for the backend, the application ensures low resource usage and offline compatibility. The system issues audio and vibrational alerts upon detecting drowsiness and logs events to promote safer driving habits. Achieving a validation accuracy of 96.03% for the EAR/MAR model, the DDD System offers a practical, scalable solution for reducing road accidents.