Drivers Drowsiness Detection using Image Processing and I-Ear Techniques

Abstract

Fatigue and micro-sleep at the wheel are often the cause of serious accidents. Consequently, the initial signs of micro-sleep can be detected before a critical situation arises. Drowsiness detection technique helps to prevent accidents caused by the drowsy drivers. Driver drowsiness detection involves three methodologies namely face detection, eye detection, drowsiness detection. Face detection technique is used to detect the driver’s face by extracting the facial features using Haar Cascade Classifier Algorithm also it detects the person’s eye region. This paper proposed to Identifying Eye Aspect Ratio (I-EAR) method to identify the eye closure of vehicle’s driver. If the driving person’s eye is closed for minimum of 40 frames or eye blink ratio is less than 8 per minute, then the proposed system identifies that the vehicle driving person is drowsy. Consequently, it began to alert sound intimation that vehicle driver is drowsy. This alert may help the drivers as a wakeup call to take an immediate action by taking rest/refreshment and continue to drive safely.