Automated Consensus-Based Data Verification in Caltrans Detector Testbed

Abstract

A practical need to assess the accuracy and attributes of each of the many types of roadway sensors and detectors motivated the California Department of Transportation to construct a traffic detector test bed on I-405 in Southern California. With up to 10 detectors of different types under concurrent test in each of six lanes, a means for automating the testing process became imperative because traditional human verification methods were not practical. This paper describes an automated data acquisition and verification system that uses a consensus of the results from all detectors under test, along with those of a reference image processing system, to create a preliminary ground truth record requiring manual verification for only a small percentage of ambiguous cases. Individual detector performance is then assessed by comparison with this verified data set. The test bed architecture, data fusion methodology, computer vision methods, operator interface, and performance results are discussed.