Office for Technology Commercialization
http://www.research.umn.edu/techcomm
612-624-0550

Intelligent Transportation System with Vision-based Tracking, Detection and Classification

Technology #z01178

Vision-Based Traffic Management Software

The vision-based vehicle tracking, detection and classification software analyzes video images, detects and counts vehicles, and tracks their movement. This intelligent transportation system identifies and classifies discrete shapes such as buses, trucks, cars and bicycles, and keeps count of the varying shapes. The traffic management software identifies where multiple lanes of traffic must cross to enter or exit (weaving sections) and merging patterns, provides an understanding of traffic capacity, and maximizes traffic patterns in areas of merging and cross-lane traffic. The vision-based tracking software was developed for the Minnesota Department of Transportation in order to acquire data for highway weaving sections in the Minneapolis and St. Paul metropolitan area.

MN-IP Try and Buy
Try
  • Trial period of 6 to 12 months. $5000/6 months.
  • Fee waived for MN-based companies or if sponsoring $50,000+ in research.
Buy
  • Exclusive license for a $15,000 conversion payment.
  • No patent costs.
  • Royalty rate of 2% (1% for MN company) after first $1 million in product sales.

** View the Term Sheet **
** Contact Andrew Morrow for more information.

Intelligent Transportation System Provides Monitoring of Weaving Sections

Road traffic control and monitoring at weaving sections is a challenging problem since vehicles are crossing paths, changing lanes, or merging with traffic as they enter or exit a highway. The vision-based traffic management software will detect vehicles and track them as they change lanes. It provides the velocity and the direction of each vehicle in the section of roadway, and comprehensive data about the section being monitored. The software works for a variety of highway layouts under various weather conditions. The system uses a single fixed camera mounted in an arbitrary position. The cameras are much less disruptive to install than loop detectors which are embedded in the asphalt.

Intelligent Transportation System Alleviates Obstruction Issues

The system uses simple rectangular outlines to represent vehicles as they move. Overlaps among vehicles and obstructions are solved by allowing vehicle outlines to overlap in the image space and by maintaining their existence even when obstructed from view.

Intelligent Transportation System Uses Vehicle Classification

Vehicle classification is an inherent problem for traffic management software. Given the wide variety of shapes, sizes and categories of vehicles, simple parameters are not sufficient to accurately categorize vehicles. The vision-based tracking system successfully classifies vehicles into various categories after they have been detected. The system identifies discrete shapes such as buses, trucks, cars and bicycles, and keeps traffic counts of the varying shapes. Vehicle classification is important in understanding the use of individual streets and highways or use of specific roadways by select vehicle classes. The system can be used to provide data for individual design scenarios (e.g. lane-specific pavement thickness).

FEATURES OF TRAFFIC MANAGEMENT SOFTWARE:

  • Data-collection for weaving sections
  • Optimal road traffic control at weaving sections
  • Real-time traffic simulation
  • Traffic prediction