System for Detecting Driver Alertness while Driving

Abstract

In this project, we are going to present a system for sleep detection alarm to monitor the driver, based on the real time surveillance and alert him as well as post it at remote location whenever it’s necessary using cloud platform. This device is to be developed using the Raspberry Pi, Open CV library and camera module. The required coding part of the project will be done using Python language. The main component of the project will be pretrained landmark detector as a software part. It identifies 68 points on the human face. The Dlib’s landmark will detect 68 facial landmarks which enables us to extract the various facial structures using simple Python array slices. The facial landmarks of fully closed eye and a fully opened eye will be first plotted. This data is further processed and tested with some results which will give the information about driver’s alertness. Once the facial landmarks associated with an eye are determined, we can apply the Eye Aspect Ratio (EAR) algorithm. In our case, we’ll be monitoring the eye aspect ratio to see if the values of the facial landmarks, thus implying that the driver/user has closed their eyes or distracted from driving or yawn. Once implemented, our algorithm will start by localising the facial landmarks on real time basis. We can then will be able to monitor the eye aspect ratio to determine if the eyes are close or nearly close which will be the indicator for driver is falling asleep. And then finally raising an alarm if the eye aspect ratio is below a pre-defined threshold for a sufficiently long amount of time. The alarm will be loud enough to wake up the driver and bring back his attention. At the same time data is passed to remote location using cloud whenever it’s necessary.