Real-Time Smart Drivers Drowsiness Detection Using DNN

Abstract

Drowsiness has become the biggest problem in the peoples life which results to ineffective work or traffic accidents. There will be huge loss to the property as well as to the lives of the people due to drowsy driving. Therefore, a real-time smart drivers' drowsiness detection system is developed using DNN. The main aim of the project is to detect and analyze the face structure and objects in the frame. Viola Jones and YOLO algorithms are used for detection of face and objects in the frame respectively. Once the face and object gets detected then the movement in the eye is analyzed. PERCLOS is used for calculation of Eye Aspect Ratio (EAR). When the EAR value is less than the threshold value then alert is given to the driver similarly alert will be triggered when there is an object in the frame by using YOLO algorithm. The real-time experimental results shows that the proposed method is highly accurate and advanced in detection of drowsiness and identification of objects in the frame.