Around the world, car accidents claim more lives with each passing year. In fact, according to the Department of Road Transport and Highways in India, the number of auto accidents increased by 34 percent from 1994 to 2004, with one-quarter of crashes due to driver distraction or falling asleep at the wheel. To put the brakes on this deadly trend, engineers at the Indian Institute of Technology innovated with NI LabVIEW software and an NI Compact Vision System to create a real-time, reliable method to detect early symptoms of driver fatigue and prevent car accidents.
The most effective estimator of a driver’s fatigue level is the measure of percentage closure of eyes (PERCLOS), followed by the rate of eye saccades, or quick, simultaneous movements of both eyes in the same direction. As a driver grows drowsy, the rate of eye saccades slowly decreases.To power the driver attention monitoring system, engineers turned to LabVIEWand NI CVS-1456 to implement algorithms for real-time eye detection and PERCLOS and eye saccades measurement. These CVS devices are easy-to-use, real-time imaging systems that acquire, process, and display images from IEEE 1394 cameras, which are then saved using NI-IMAQ driver software.
Offline results for eye detection and eye state estimation based on algorithm in NI Vision
The final result? A nonintrusive, stand-alone embedded system that achieved 6 fps with an accuracy of more than 90 percent on the CVS-1456 device. This driver attention monitoring system may not be a guardian angel, but it’s just another way that NI tools are helping the world put safety first.
>> Read the full case study here.