Driver Drowsiness Detection System

Abstract

Drowsiness is one of the major causes of driving accidents, and it results in several road fatalities. These days, driver drowsiness is one of the prime causes of most of the accidents in the world. To solve such a problem, a driver drowsiness detection system is developed, which uses the closure period of eyes to detect drowsiness. Especially the system continuously analyses the movement of eyes of the drivers. When the duration of the eyes is not normal, an alarm gets activated to warn or alert them. The system is implemented on python with a single camera. Different machine learning libraries are used such as Keras, Cv2, OS, Matplotlib, Tensorflow and Numpy. Deep Neural Network is used to demonstrate the system's good performance in terms of accuracy in drowsiness detection results and thus reduces road accidents. This project aims to use image processing techniques to detect the driver's drowsiness in a driving simulator and reduces road fatalities.