AUTOMATED DRIVER ALERTNESS MONITORING SYSTEM USING EYE BLINK DETECTION

Abstract

This project introduces a real-time driver drowsiness monitoring solution that identifies signs of fatigue through visual cues like extended eye closure and yawning. Utilizing a standard webcam, the system continuously observes facial landmarks with a focus on eye activity. By calculating the Eye Aspect Ratio (EAR), the system can determine whether the driver is drowsy based on eye behavior. This detection leverages Dlib and MediaPipe libraries for facial landmark analysis, enhancing both accuracy and efficiency. Once signs of drowsiness are detected, a visual alert is displayed on the screen to prompt driver awareness. The system is implemented in Python, using OpenCV for image processing and Tkinter for an intuitive graphical interface. It operates seamlessly in real-time and does not require the user to wear any hardware, making it non-intrusive and user-friendly. Looking ahead, incorporating machine learning techniques could make the detection more robust across different user profiles and environmental conditions.