A NEW DRIVER DROWSINESS AND DETECTION SYSTEM

Abstract

Road accidents caused by driver fatigue and drowsiness have become a major concern in modern transportation systems, leading to significant loss of life and property worldwide. Continuous driving, lack of rest, and poor alertness levels are among the primary reasons that contribute to drowsy driving. To address this critical issue, this project presents the design and implementation of a Driver Drowsiness Detection System using real-time computer vision techniques.The proposed system is developed using Python and leverages powerful libraries such as OpenCV and Dlib to monitor the driver’s facial features through a live video feed captured by a webcam. The system utilizes Dlib’s 68-point facial landmark detector to accurately identify key facial regions, particularly the eyes. By analyzing the eye landmarks, the Eye Aspect Ratio (EAR) is computed in real time to determine the driver’s alertness state. A sustained decrease in the EAR value indicates eye closure, which is used as a primary indicator of drowsiness or sleep.The system employs time-based logic to differentiate between normal blinking, drowsiness, and sleep conditions. When drowsiness or prolonged eye closure is detected, the system immediately triggers visual warnings and audio alerts to alert the driver and help prevent potential accidents