Using Individual Calibration to Make a More Accurate Active Infrared Drowsiness Monitor

Abstract

This project aims to create an improved wearable drowsiness monitor for eventual deployment on glasses or goggles. A noninvasive drowsiness monitor has the potential to help users make smarter decisions about when to stop dangerous activities like driving, or to empower organizations to regulate the behavior of drowsy individuals. The project's unique approach combines active infrared sensing to measure spontaneous blinks, whose features indicate a level of drowsiness, and calibration of these blink features to a user. In an exploratory research subject population, this project presents the potential for such a system to improve drowsiness detection accuracies and sets the stage for future validation in a larger study.