Automated Video-based Traffic Count Analysis

Source Organization:
Principal Investigator:C.J. Taylor
PI Contact
Project Manager:
Funding Source(s) and Amounts Provided (by each agency or organization):
Total Dollars:
Agency ID/Contract/Grant Number:
Start and End Dates:
Project Status:Complete
Subject Categories:transportation; operations and traffic management
Abstract:The project goal is to develop freely available software to automate the tedious task of analyzing video data to produce traffic counts. Traffic count is collected by transportation planners and used in the design, deployment and maintenance of safe roadways. Good data are critical to good decision-making. Existing commercial systems like Miovision offer solutions but are expensive and somewhat cumbersome. Using a virtual equivalent of a network of pneumatic tubes, we will analyze the data to determine vehicle traffic patterns. We define a set of lines in the image as shown and then detect when these lines have been ‘crossed’. By analyzing the relative timing of event sets, we discern when and where vehicles enter and leave the intersection.
Describe Implementation of Research Outcomes (or why not implemented):
Impacts/Benefits of Implementation (actual, not anticipated):
Project URL: