, Professor, Computer Science and Engineering
Collisions between vehicles at urban and rural intersections account for nearly a third of all reported crashes in the United States. This has led to considerable interest at the federal level in developing an intelligent, low-cost system that can detect and prevent potential collisions in real time. Monitoring traffic intersections in real time as well as predicting possible collisions is an important first step towards building an early collision warning system. This project studied general vision methods used in a system addressing this problem, including practical adaptations necessary to achieve real-time performance. Specific techniques and methods developed by this research include a novel method for three dimensional vehicle size estimation; a method for target localization in real-world coordinates, which allows for sequential incorporation of measurements from multiple cameras into a single target's state vector; a fast implementation of a false-positive reduction method for the foreground pixel masks is developed; and, a low-overhead collision prediction algorithm using the time-as-axis paradigm. The prototype system developed in this research proved able to perform in real-time on videos of quarter-VGA (320x240) resolution.
- Project number: 2002025
- Start date: 03/2003
- Project status: Completed
- Research area: Transportation Safety and Traffic Flow
Safety, Traffic operations