Facial Detection for Recognition of Drowsiness

Abstract

Drowsiness and fatigue among drivers are major contributing factors to road ac- cidents. These critical problems can damage road facilities, vehicles, and, most notably, human losses. This is due to the driver’s fatigue, and the causes of his fatigue are multiple, such as lack of sleep, long journey, insomnia, alcohol consumption, and mental pressure. Each of which can lead to serious disaster,The current transportation system needs to be improved for more road dangers. Thus, more accidents can be avoided by incorporating automatic fatigue detection systems into vehicles. The drowsiness detection system continuously analyzes the driver’s level of alertness and alerts the driver before the arrival of a dangerous threat to safety. This mechanism is suitable for alerting the driver to stay awake throughout the journey. Thus, this study proposed a real-time prototype for recognizing drowsiness and fatigue. This study includes face detection with two methods for detecting the face in the designated digital image : feature-based and image-based approaches. Then we dealt with determining the exact location of the left and right eyes by checking the eye using devices SVM. Then we used the information on the eye contour to discover Whether the eye is open or closed after determining the location of the eyes. The proposed method is powerful for dealing with light changes, mild circling, wearing glasses, and partial facial occlusion. The proposed method is evaluated on the BioID facial database.