Monitoring and Predicting Pedestrian Behavior at Traffic Intersections

Most traffic intersections lack awareness of pedestrian traffic: their perception abilities—when available—are usually limited to the detection of vehicles at very specific places. Video cameras can be used to monitor pedestrian traffic in a setting where a static camera that has an unobstructed view of the road is used to detect and track pedestrians. However, simply detecting a pedestrian is not enough to fully describe the motion of that person within the context of the intersection: it is also necessary to determine where that person is located. The correct placement of a person within the intersection requires both the intrinsic and extrinsic calibration of each camera, so that geometric locations on the surfaces where pedestrians move can be transformed to pixels in each camera’s image and vice versa. Traditional approaches to camera calibration work well for the intrinsic part, and can be conducted before the cameras are mounted. However, the extrinsic calibration part requires elements that are not well suited for this application (e.g. collection of multiple images showing a calibration target at different locations), and are time consuming. This study intends to reduce the time and amount of effort needed to calibrate and set up a video camera at any location. In this work the research team focused on developing an approach to calibrate traffic cameras that simplifies the extrinsic calibration work on-site using a low-cost custom-made laser scanner, and that requires minimal personnel training.

• Record URL:
  • http://ppms.cit.cmu.edu/media/project_files/065_finalReport.pdf
• 
• Summary URL:
  • http://ppms.cit.cmu.edu/projects/detail/65
• Supplemental Notes:
  • This document was sponsored by the U.S. Department of Transportation, University Transportation Centers Program.
• Corporate Authors:

  Carnegie Mellon University

  Robotics Institute, 5000 Forbes Avenue
  Pittsburgh, PA  United States  15213-3890

  Technologies for Safe and Efficient Transportation University Transportation Center

  Carnegie Mellon University
  Pittsburgh, PA  United States  15213

  Office of the Assistant Secretary for Research and Technology

  University Transportation Centers Program
  Department of Transportation
  Washington, DC  United States  20590
• Authors:
  • Navarro-Serment, Luis E
• Publication Date: 2016

Language

• English

Media Info

• Media Type: Digital/other
• Edition: Final Research Report
• Features: Figures; Photos;
• Pagination: 8p

Subject/Index Terms

• TRT Terms: Behavior; Calibration; Data collection; Intersections; Lasers; Pedestrian detectors; Pedestrian movement; Pedestrians; Scanners; Traffic surveillance; Video cameras
• Subject Areas: Data and Information Technology; Highways; Operations and Traffic Management; Pedestrians and Bicyclists;

Filing Info

• Accession Number: 01679719
• Record Type: Publication
• Files: UTC, TRIS, ATRI, USDOT
• Created Date: Aug 30 2018 9:43AM