Intelligent Transportation Systems: Intersection Monitoring

Artificial Intelligence, Robotics, and Vision Lab
University of Minnesota, Minneapolis, Minnesota

Detection and Classification of Vehicles

Overview

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. Our research is aimed at the development of a system that uses video cameras to continuously gather traffic data at intersections (e.g., vehicle speeds, positions, trajectories, accelerations/decelerations, vehicle sizes, signal status etc.) which might eventually be used for collision prediction.

At present, the major thrusts of our research are:

developing robust, efficient position estimators to track vehicles
integrating data from multiple cameras to yield more accurate vehicle position estimates
creating a system for modeling intersection activity and predicting collisions using computational geometry

Publications:

A Vision-Based Approach to Collision Prediction at Traffic Intersections. (PDF): S. Atev, H. Arumugam, O. Massoud, R. Janardan and N. Papanikolopoulos. A Vision-Based approach to Collision Prediction at Traffic Intersections. IEEE Trans. on Intelligent Transportation Systems, vol. 6, no. 4, pp 416-423, December 2005.
A Collision Prediction System for Traffic Intersections. (PDF): S. Atev, O. Masoud, R. Janardan and N. Papanikolopoulos. A Collision Prediction System for Traffic Intersections. In Proc. IEEE/RSJ Int'l Conf. Intell. Robots and Syst. (IROS 2005), pp. 2844−2849, Aug. 2005.
Computer Vision Algorithms for Intersection Monitoring. (PDF): Harini Veeraraghavan, Osama Masoud, N. P. Papanikolopoulos. Computer Vision Algorithms for Intersection Monitoring. IEEE Trans. on Intelligent Transportation Systems, vol. 4, no. 2, pp. 78-89, Jun 2003.
A Real-Time Collision Warning System for Intersections. (PDF): Kristen Stubbs, Hemanth Arumugam, Osama Masoud, Colin McMillen, Harini Veeraraghavan, Ravi Janardan, and Nikos Papanikolopoulos. A Real-Time Collision Warning System for Intersections. Intelligent Transportation Systems America. May 2003.
Vision-based Monitoring of Intersections. (PDF): Harini Veeraraghavan, Osama Masoud, and Nikolaos Papanikolopoulos. Vision-based Monitoring of Intersections. In Proc. Intelligent Transportation Systems Conference. September 2002.
Real-Time Tracking for Managing Suburban Intersections. (PDF): Harini Veeraraghavan, Osama Masoud, and Nikolaos Papanikolopoulos. Real-Time Tracking for Managing Suburban Intersections. In Proc. Digital Signal Processing. July 2002.

Demos:

Vehicle and Pedestrian Tracking Movie (3.4M)
This clip shows some of the tracking results for a traffic scene. The colored arrows indicate the direction of motion.

Collision Detection Simulation (2.1M)
This is a simulated traffic scene viewed from above. A red line between two vehicles indicates that the vehicles are too close or are projected to be too close.

Collision Detection Applied to a Real Traffic Scene (4.4M)
This clip shows how the tracking module is able to correctly detect vehicle sizes. It also shows the output of the collision prediction module which gives advance warning (one second ahead) of likely collisions.

People:

Faculty Nikolaos Papanikolopoulos Ravi Janardan	Research Associate Osama Masoud
Graduate Students Stefan Atev Harini Veeraraghavan	Undergraduate Students Grant Miller

This work is supported by grants from the National Science Foundation, Minnesota Department of Transportation, and the Intelligent Transportation Systems Institute.
Any opinions, findings, and conclusions or recommendations expressed in this material are those of the author(s) and do not necessarily reflect the views of the National Science Foundation.