Enhanced Driver Safety through Real-Time Risk Assessment

Created using ChatSlide

This project presents a real-time vision system designed to enhance driver safety by detecting risky behaviours such as eye closure, yawning, and phone usage. It evaluates the driver's state as SAFE, WARNING, or DANGEROUS to reduce accidents due to drowsiness. The system uses a webcam-based detection approach with modules for eyes, yawns, phone usage, and posture, employing machine learning and YOLOv8 for accurate detection. It incorporates temporal logic to minimise false alerts and...