Driver Drowsiness Detection Using Haarcascade Algorithm

Abstract

Recently, in addition to research and development of autonomous vehicle technology, machine learning systems have been used to assess driver positions and emotions to upgrade road safety. The driver's position is assessed not only by basic characteristics such as gender, age, and driving experience, but also by the driver's facial expressions, biosignals, and driving behavior. Recent developments in video processing using machine learning have made it possible to analyze images obtained from cameras with high accuracy. Therefore, based on the relationship between facial features and the driver's sluggish position, variables that reflect facial features are established. In this paper, we propose a method of collecting detailed features of the eyes, mouth, and head using the OpenCV and DeLib library to assess the driver's level of drowsiness.