Driver Drowsiness Detection using OpenCV and DLib

The proposed system aims to enhance transportation safety by reducing the occurrence of accidents caused by driver fatigue and drowsiness, a prevalent issue in recent times. To detect signs of drowsiness and fatigue in drivers, various facial and body gestures are analyzed, including tired eyes and yawning. Our system uses multiple features such as blinking and yawning frequency and the Eye Aspect Ratio(EAR) to calculate the ratio of distances between the distance between the eyelids and the length of one eye and locates 68 key points which is crucial for detecting drowsiness. Similarly, Yawning is detected by calculating the distance between the upper and lower lip and comparing it to a threshold value. Our proposed system achieving an accuracy of over 90 % for all tested cases of lighting conditions and spectacles is expected to reduce the number of accidents and contribute to the advancement of technology aimed at reducing fatalities resulting from road accidents.