Eye-Com Research Validates PERCLOS for Drowsiness Detection: Part 1

This is the first in a series of five posts detailing the research conducted by Eye-Com Corporation in cooperation with the U.S. Department of Defense and the U.S. Department of Transportation.

Part 1: Background

Startling Statistics Present a Need For Driver Drowsiness Detection Technology

Of the many public safety issues plaguing our roadways today, drowsy-driving is of major concern. Those who drive while sleep-deprived, whether it’s a result of lifestyle choices or because of clinical sleep disorders, are shown to perform as poorly behind the wheel as legally intoxicated drivers.

According to the National Sleep Foundation (NSF), more than one-half of adults have admitted to driving while drowsy at least once in the past year. These numbers are unlikely to improve; NSF polls report that more and more Americans are getting inadequate sleep. Chronic sleep deprivation is especially common amongst shift workers, who account for 20% of the US work force and have twice the number of highway accidents as day workers. Thirty one percent of fatal commercial accidents are drowsiness related, causing the 1995 US Truck and Bus Summit to identify driver fatigue as its number-one safety issue. Driver fatigue is estimated to result in 1500 fatalities, 70,000 injuries, and over $54 billion in economic loss, health care costs, and loss of productivity each year.

Due to these alarming statistics, many studies have been conducted over the past fifteen years to identify what technology is the most effective at detecting and predicting driver fatigue. In 2000, Krueger Ergonomics Consultants conducted a comprehensive review of the available technologies in an effort to determine which had the most potential for future development. Krueger concluded that the ideal technology would, amongst other criteria, be able to perform in real time, be non-obtrusive to the driving process, be operational under all lighting conditions, be reliable and sensitive enough to detect even slight loss of attention or fatigue, and be able to predict the operator’s future state.

Possible technologies that met these criteria included head-movement detectors, EEG or EKG algorithms, and various eye-related technologies. Of the eye-related technologies, those that measure PERCLOS are considered to be one of the most reliable in determining a driver’s alertness levels.

To be continued in Part 2: The PERCLOS Measure….

Comments