Computer Program Warns Weary DriversPosted July 28th, 2009 by Spencer Chin
Although less publicized than alcohol, driver fatigue is a major contributor to auto accidents in the U.S. The National Highway Traffic Safety Administration (NHTSA) estimates that 56,000 sleep-related road crashes occur annually in the U.S., resulting in 40,000 injuries and 1,550 fatalities. According to the National Transportation Safety Board, 52 percent of 107 single vehicle incidents involving heavy trucks were fatigue-related, confirming the belief that fatigue is a significant problem for long-distance truck drivers.
Unfortunately, no standard screening method exists to detect weary drivers, as breathalyzers exist to detect drunk drivers. But there may be hope in a computer program developed by researchers at Vanderbilt University, Nashville, TN, and two India-based universities.
Using an in-car camera, the program uses image processing software to capture a sequence of images on the driver’s face. Analyzing facial expression changes, the program identifies yawning as a facial movement distinct from other movements, such as smiling, talking, and singing.
The researchers say the algorithm is effective at detecting yawns, regardless of image intensity and contrast, small head movements, viewing angle, spectacle wearing, and skin color. The program correlates yawn frequency with fatigue behavior and could be hooked up to a warning system to alert drivers.
Although it may be years before such a system is implemented in vehicles, I would welcome it sooner based on my own experience with driver fatigue. Earlier this year, I nodded off on an interstate highway in New Jersey and meandered onto the side of the road. My car clipped a guardrail, damaging the right front quarter panel and requiring replacement. Fortunately, I was not injured.
Driver fatigue can happen to anyone. I would welcome any means to warn drowsy drivers to take a much-needed break.