Abstract: Driver drowsiness is a major contributor to road accidents worldwide, leading to numerous injuries and fatalities each year. This project aims to develop a real-time drowsiness detection system using Python and Google's MediaPipe framework. By analyzing facial landmarks, particularly the eyes, the system can detect signs of drowsiness such as frequent blinking or prolonged eye closure. The Eye Aspect Ratio (EAR) is calculated to monitor eye state, triggering alerts when drowsiness is detected. This low-cost, non-invasive solution enhances road safety by preventing accidents due to driver fatigue. The project emphasizes real-time processing, ease of deployment, and accuracy.

Keywords: Drowsiness detection, MediaPipe, Python, Eye Aspect Ratio (EAR), Driver safety, Realtime monitoring, Facial landmarks, Computer vision, OpenCV.

Downloads: | DOI: 10.17148/IARJSET.2025.1211030