Drowsiness Detection using Fuzzy Inference System

Abstract

In this paper, a prototype of drowsiness detection system is developed to reduce car accidents and save drivers' life. In such real-time system, the state of driver' eyes is taken continuously and sending alert if required. A low cost system for detecting a drowsy condition of a driver of a vehicle includes a video imaging camera located in front of the driver in the vehicle oriented to generate images of a driver of the vehicle. The image algorithm processes the acquired image. The algorithm of Driver Drowsiness Detection System depends on Fuzzy Inference System comprises the steps of binarizing the driver image from camera, preprocessing and extracting eye.The processor monitors an eye and finds whether eye is in an opened or in a closed state by finding relative distances between eyes corners. The processor further determines driver drowsiness condition and alerts the driver.Our system was implemented online and in a real car using embeeded system that facilates the Rasperry Pi 3 as the main controller and trained with sample images captured for chosen human groups that mimic the drowsy peoples drivers and sleepy drivers for system evaluation.The final results was robust and the accuracy of alarming the sleepy driver was god and sufficient and the accuracy of detection was nearly 95%.