Eye Tracking System to Detect Driver Drowsiness using Deep Learning

Abstract

An insufficient amount of sleep regimen can have an enormous impact on your quality of life. According to research, being subjected to stress at work, doing too much on the laptop, being on your smartphone, and experiencing problems with sleep deprivation. At the same time, driving can double the chances of you being tired behind the wheel. They are fatigued and drowsy while driving is a few of the reasons why there are more traffic accidents. Often, as a result of mental or physical exhaustion, people can fall asleep and face difficulties. This thesis discusses a method for determining whether a driver is sleepy behind the wheel and helps the person stop an accident. As a goal, one side effect of this initiative’s overall goal is to cut traffic accidents. We want to boost drivers’ alertness and make people’s attention span longer. Masking drowsiness while driving might lead to frequent yawning and drooping of the eyelids, or getting progressively drowsier and start to fall asleep behind the wheel, might occur. We used the network-based face and eye-expansion feature extraction algorithm to identify the driver and locate his pupils. To calculate the percentage of eyelid closure over time, we use the driver’s eye closing characteristic