Driver Alertness Identification And Alarming System Using Haar Cascade Algorithm

Abstract

Motorist fatigue has been one of the top causes of automobile accidents throughout the world in recent years. The state of the driver, i.e. drowsiness, is a simple way of determining driver fatigue. It is vital to recognise the driver's tiredness in order to protect lives and property. The purpose of this project is to construct a prototype of a drowsiness detection system. This is a real-time system that continuously captures photos and analyses the eye's condition using the approach described, as well as delivering warnings as required. Although various methods for assessing fatigue exist, this technique is fully non-intrusive and so has no effect on the driver, revealing the driver's genuine state. The retina's per-closing value is utilised to detect whether or not a person is tired. When a driver's eyelids close more than a particular amount, he or she is deemed drowsy. This system is made up of numerous OpenCv libraries, the most important of which being Haar-cascade. Furthermore, to improve the driver's security, as well as to check if the driver is adhering to the "do not drunk and drive" rule. Before the automobile starts, the amount of alcohol is detected, and if the driver is determined to be drunk, the automobile will not start. This keeps the driver out of trouble while simultaneously keeping him safe.