Real-Time Driver Drowsiness Detection and Alert System

Nowadays, road accidents have become really high and is causing severe physical injuries, deaths etc, mainly in India. India has become the highest in the world in case of road accidents, recording 53 road crashes per hour. Drowsy driving is one of the reasons for road accidents. Increasing road accidents due to drowsy driving indicate the need of system that detect the drowsiness of the driver and alert them at the correct time. “Researchers have attempted to determine driver drowsiness using the following measures: (1) vehicle-based measures; (2) behavioral measures and (3) physiological measures”[13]. Rate of using physiological measures to detect drowsiness is high; vehicle-based measures are affected by construction of the street, driving ability of the driver, type of vehicle etc; some methodologies use mental measures for which some anodes are put on the head and body. So here we have used much more feasible method. In this paper we discuss a method which is non-intrusive. In our proposed system, the main data we use to detect driver drowsiness is the eye conclusion proportion, its duration etc. Eyes opening and closing ratio reflects a person's mind status and attention and therefore, it can be potentially used to indicate driver fatigue levels[14]. We keep a value as the threshold value, if the eye conclusion proportion goes below the threshold value, then we alert the driver using a buzzer and provide tips to get rid of the drowsiness through an audio message. For continuously detecting the driver’s eye we use a camera. We also use an LED light to alert the co-passengers about the drowsiness of the driver.