Drowsiness Detection And Alert System Using Deep Learning

Abstract

A low cost and simple method for detecting the eye closing of the driver by using deep learning while driving. The concept of deep learning uses the convex hull function to cascade the eye separately from the face. The eye aspect ratio is used for the detection of closing of the eye. If the eye aspect ratio value is less than the threshold value, the counter starts to increase until the driver opens the eye. The notification will be displayed on the mobile application for alerting the driver, if the driver fails to response the given notification the speed will be reduced. Once the counter reaches the threshold value the motor will gets stopped completely and the message will be send to the given number.