D-Cubed: Driver Drowsiness Detection

The U.S. National Highway Traffic Safety Administration has indicated that driving while drowsy is the cause of 22 to 24 percent of car crashes, and results in a four- to six-times higher crash/crash risk compared to driving while alert. Therefore, the use of assistive systems that monitor a driver’s level of vigilance and alert the driver in case of drowsiness can be significant in the prevention of accidents.  In this project, I will study, design, and implement a system to detect drowsiness by recognizing yawning from mouth geometry and facial movements, based on CogniVue APEX™, which is a programmable embedded platform offering multi-camera solutions installed in cars.

Behnoosh Hariri
Faculty Supervisor: 
Dr. Shevin Shirmohammadi