Improved Eye Tracking Algorithm to Be Used for Driver Fatigue Detection System
Researchers from the Department of Computer Science and Information Engineering at Tamkang University in Taipei, Taiwan have found an improved eye tracking algorithm, which they plan to use in detecting driver fatigue.
The department has proposed a vision-based, real-time driver fatigue detection system that uses ordinary CCD cameras to accomplish a variety of tasks: face detection, eye detection, eye tracking, and fatigue detection.
The researchers conducted an experiment aimed at improving the eye tracking technique utilized by the system. The results revealed that the fast search algorithms, 2D-log and three-step search, used in MPEG video coding are more efficient than the exhaustive search method originally developed to track eye movement. The original exhaustive search examined every search point within the search area to find the best possible matches. These two fast search algorithms reduce search computation by greatly decreasing the number of search points required. For example, the 2D-log search only needs about 5% of the computations required for the original exhaustive search, yet it can attain a slightly higher rate of correct tracking.
Findings such as these have brought the research team a step closer to creating an effective driver fatigue detection device using eye tracking technology.
Read more…
Related articles:
- Eye Tracking Research Delves Into Eye Detection Algorithm
- Eye-Com Eye Tracking Drive Simulator Studies Driver Inattention
- Eye tracking lie detection system 82.5% accurate
- Eye Tracking Study Finds Cell Phones Not Leading Cause of Driver Distraction
- Eye Tracking: How Smart is Your Car?
- Eye Tracking: Evaluating Teen Driver Distractions
- EyeSeeCam Portable Eye Tracking System
- Eye Tracking: Avatar Lie Detection … It’s All in the Eyes
- openEyes Eye Tracking System
- Improving Password Security With Eye Tracking