01555315

Component

Validating the accuracy of sensors is an essential step in the collection of traffic speed data. The accuracy of automated speed data has been evaluated in small- and large-scale tests using multiple technologies and methods. While inductive loops are standard, video-based detectors have demonstrated the ability to substitute conventional detection devices. Though existing literature documents several issues associated with extracting vehicle speeds from video, the analysis of speed data, especially at the microscopic or individual level, has been limited. The purpose of this paper is to evaluate the accuracy of a video-based detection system, comprised of commercially available video cameras and an open-source computer vision software system. Several camera orientations were tested along an urban arterial and a highway in Montreal, Canada. A semi-automated vehicle tracking process was used to extract the vehicle speeds, which were compared to manually observed speeds. Although the traditional mean relative error approach led to unacceptable results, a new approach was proposed for the evaluation of traffic detection technologies. The segregated error approach divides simplistic mean error into separate values for accuracy and precision. In doing so, several of the camera orientations exhibited precision error values within the accepted range for speed data quality (5%). Even with large errors, the potential exists to calibrate video-based speeds, by removing the over- or underestimation bias, to acceptable performance levels as long as precision error is minimized through appropriate selection of camera position and orientation.

This paper was sponsored by TRB committee ABJ50 Information Systems and Technology.

01550057

15-1337

English

14p

2015

Transportation Research Board 94th Annual Meeting

Digital/other

Figures; References; Tables

Accuracy; Automatic vehicle detection and identification systems; Computer vision; Performance measurement; Speed; Traffic data; Video imaging detectors

Data and Information Technology; Highways; I72: Traffic and Transport Planning

Transportation Research Board Annual Meeting 2015 Paper #15-1337

TRIS, TRB, ATRI

Dec 30 2014 12:31PM