DROWSY DRIVER DETECTION WITH CNN AND RNN

Abstract

This project is focused on drowsy driver detection and the objective of this project is to recognize driver’s state with high performance. Drowsy driving is one of the main reasons of traffic accidents in which many people die or get injured. Drowsy driver detection methods are divided into two main groups: methods focusing on driver’s performance and methods focusing on driver’s state. Furthermore, methods focusing on driver’s state are divided into two groups: methods using physiological signals and methods using computer vision. In this project, driver data are video segments captured by a camera and the method proposed belongs to the group that uses computer vision to detect driver’s state. There are two main states of a driver, those are alert and drowsy states. Video segments captured are analysed by making use of image processing techniques. Face is localized in the image and key facial structures (Eye, Mouth) are detected on face ROI(region of interest). Then eye aspect ratio(EAR) and mouth aspect ratio(MAR) is calculated for horizontal and vertical distances. If EAR is less than some threshold value then eye is detected as closed and if MAR is greater is than some threshold value then yawning is detected. As closed eye and yawning are two most symptoms of drowsiness, we can easily predict the driver’s state is alert or drowsy based on these two observations.