CAMERA POSITIONING AND CALIBRATION TECHNIQUES FOR INTEGRATING TRAFFIC SURVEILLANCE VIDEO SYSTEMS WITH MACHINE-VISION VEHICLE DETECTION DEVICES

This project addresses two fundamental areas. First, the project evaluates the technical difficulties associated with integrating machine vision vehicle detection with existing VDOT CCTV camera and camera control equipment (such as calibration, repositioning, and data loss due to pan/tilt operations by traffic controllers). Second, the project develops automated techniques to address these issues. The development effort concentrates on creating repositioning and recalibrating techniques with machine vision technologies. This project develops a prototype approach to integrate VDOT CCTV equipment with existing machine vision traffic detection devices. The prototype and the experience gained developing the prototype, will improve VDOT's understanding of machine vision traffic detection technologies and their appropriateness for system integration in traffic management applications; provide solutions to typical problems associated with machine vision systems in traffic control centers; serve as the "proving grounds" for the development of machine vision integration plans; and potentially increase the return on VDOT investments in CCTV systems