DISTRACTED DRIVER DETECTION USING CONVOLUTIONAL NEURAL NETWORK

Abstract

The major cause of accidents is due to driver’s distracted actions. Most of the car accidents involve driver distraction under different forms such as talking on the phone, texting, operating the radio, drinking and talking with the passenger and so on. In most cases of distractions, a driver just keeps only one hand or even no hand on the steering wheel. Therefore, detecting driver’s distracted behavior aims at several goal, i.e. the levels of pay attention of a driver to the road and using two hands or not while driving. In this system, deep learning method is used to detect and classify the driver’s action. This system is developed by using Fine-Tuning AlexNet Convolutional Neural Network to train and classify the driver’s distracted behaviors. This system is also implemented by Python programming language.