Driver Fatigue Prediction with Real Time Driver Alert System Using Deep Learning

Deep learning is a type of machine learning which imitates instruction given by human. The main theme of our project is to detect whether the person was normal or abnormal while driving the vehicle, many of the accidents are caused due to the driver may be drunken or tired or feel sleepy while driving the vehicle. The proposed model will only consider the facial features like eye, nose, and lips of the intended person and detect the drowsiness by taking the eye blink duration and head posture recognition and also eye and mouth state recognition. And then detecting he/she is good to drive the vehicle or not and finally, If that person was abnormal in the sense, three different types of alert will be provided, they are self alert, message alert, and mail alert. Here, Convolutional Neural Network [CNN] algorithm is used for implementation. Convolutional Neural Network [CNN] is a type of artificial neural network, which is used in the process of image recognition and processing. It is specifically designed to process the pixel data. The proposed algorithm can be used to detect the drowsiness of the driver while driving the vehicle.