U.S. Department of Transportation
Federal Highway Administration
1200 New Jersey Avenue, SE
Washington, DC 20590
202-366-4000


Skip to content U.S. Department of Transportation/Federal Highway AdministrationU.S. Department of Transportation/Federal Highway Administration

Federal Highway Administration Research and Technology
Coordinating, Developing, and Delivering Highway Transportation Innovations

Overview

 

Research and Development (R&D) Project Sites

 
Project Information
Project ID:   FHWA-PROJ-09-0030
Project Name:   Pedestrian Detection, Counting and Tracking Systems for Travel Surveys, Traffic Safety Systems, and Traffic Control Systems (091-FH4)
Project Status:   Completed
Start Date:  August 27, 2009
End Date:  August 20, 2010
Contact Information
Last Name:  Gibson
First Name:  David P
Telephone:  202-493-3271
E-mail:  david.gibson@dot.gov
Office:   Office of Operations Research and Development
Team:   Trans Enabling Technologies Team [HRDO-10]
Program:   Small Business Innovation Program (SBIR) (2% of Adjusted CA)
Project detail
Project Description:   Phase I would develop and demonstrate prototype software and hardware that would embody a simplified version of the system. Phase II would enhance the system algorithms and sensor as a real-time traffic survey collection device. Phase II would also address issues related to making the system at least minimally functional during rain and snow events. Researchers would conduct laboratory and field tests of improved hardware and algorithms and conduct field operational tests through summer 2010. A Go-No Go decision was made on phase II to proceed with recommendation for award.
Goals:   To determine whether it is possible to count and track pedestrians using a combination of stereo imaging and light detection and ranging (LIDAR).
Background Information:   Current pedestrian and bicycle detectors do not do an adequate job of detecting, counting, and tracking pedestrians for automated counting to support traffic surveys or for traffic control. A pedestrian/bicycle detection system using artificial intelligence algorithms for pulling out and tracking pedestrians might significantly increase the accuracy of automated pedestrian counting systems and eventually lead to more pedestrian-responsive traffic control systems.
More Information URL(s):  
  1. http://www.migmasys.com/
  2. http://www.fhwa.dot.gov/publications/publicroads/09septoct/01.cfm
Field Test:   Several hours of data collection were performed. The data were then used to develop algorithms. A separate set of data were used to test the prototype algorithms.
Test Methodology:   Conduct data collection with sensors. Develop algorithms to test different approaches.
Partners:   U.S. Department of Transportation: Research and Innovative Technology Administration (RITA); Role(s): Other stakeholder
Expected Benefits:   Improved pedestrian data for pedestrian counting and surveys and for traffic control operations and safety.
Project Findings:   It was determined that it is theoretically feasible to count and track pedestrians using a combination of stereo imaging and light detection and ranging (LIDAR).
FHWA Topics:   Research/Technologies--FHWA Research and Technology
TRT Terms:   Pedestrian Safety
Operations
Research
Stereoscopic Models
Traffic Control
Laser Radar
Pedestrians
FHWA Disciplines:   Operations
Safety
Subject Areas:   Pedestrians and Bicyclists
Research

 

Federal Highway Administration | 1200 New Jersey Avenue, SE | Washington, DC 20590 | 202-366-4000
Turner-Fairbank Highway Research Center | 6300 Georgetown Pike | McLean, VA | 22101