DrowsyDetection

Drowsy driving is an important problem that causes hundreds of fatalities each year. Because of this, programs designed to detect a driver's alertness have become more and more necessary. Machine learning models that detect when a driver is sleepy are one type that have been explored. This repository explores a computer vision approach to addressing this problem. Three models were explored, one random forest, a neural network, and a convolutional neural network. Out of these models, the random forest performed the best, followed by the neural network, and finally the convolutional neural network. Overall, the models showed some promise, but were not efficient or accurate enough to be used in a real life situation.