Real time driver drowsiness detection using a logistic-regression-based machine learning algorithm

Abstract

The number of car accidents due to driver drowsiness is very steep. An automated non-contact system that can detect driver's drowsiness early could be lifesaving. Motivated by this dire need, we propose a novel method that can detect driver's drowsiness at an early stage by computing heart rate variation using advanced logistic regression based machine learning algorithm. Our developed technique has been tested with human subjects and it can detect drowsiness in a minimum amount of time, with an accuracy above 90%.