Driver Drowsiness Detection Using Deep Learning

Project Context and Objective

This project focuses on developing a real-time driver drowsiness detection system using computer vision and deep learning techniques. The system aims to enhance road safety by automatically detecting signs of fatigue through eye state analysis. Driver fatigue is one of the major causes of traffic accidents worldwide. Detecting drowsiness in real time is essential to prevent accidents and improve transportation safety.

Approach & Methodology

• Deep Learning Model: A CNN-based classification model was developed to detect eye states (open/closed).
• Face Detection: The Haar Cascade algorithm was used to detect the face region from real-time video input.
• Real-Time Evaluation: The PC webcam was integrated to test the model in a live environment.
• Database: Our database contains 4000 images of closed and open eyes, including 600 for testing and 3400 for training.
• System Pipeline: Camera input → Face detection → Eye region analysis → CNN prediction.