The accuracy of the loop-based vehicle classification under various traffic conditions is greatly dependent upon the capability of clarifying phases or states of traffic flow. One challenge lies in identifying traffic phases using variables that could be directly calculated from dual-loop data. In this paper, we present a hybrid method that incorporates level of service and K-means clustering methods for identifying traffic phases from dual-loop data. We apply the "phase representative variables" to represent traffic characteristics in the traffic flow phase identification algorithm. By using video vehicular trajectory data, we have successfully identified the traffic factors influencing vehicle classification accuracy under non-free traffic conditions and developed the innovative length-based vehicle classification models. The model evaluation results indicate that the accuracy of the estimated vehicle lengths has been increased from 42% to 92% under synchronized and stop-and-go traffic flow phases. The results also foster a better understanding of the traffic stream characteristics and associated theories to help further development of relevant microscopic simulation models with sensing traffic data sources.
Clarifying Traffic Flow Phases for Vehicle Classifications Using Dual-Loop Data
The Twelfth COTA International Conference of Transportation Professionals ; 2012 ; Beijing, China
CICTP 2012 ; 3441-3450
2012-07-23
Aufsatz (Konferenz)
Elektronische Ressource
Englisch
Traffic counts and classifications: distributed data processing
Tema Archiv | 1989
|Analyzing Traffic Flow Phases Using ITS Data
British Library Conference Proceedings | 2005
|Improved vehicle classification from dual-loop detectors in congested traffic
Online Contents | 2014
|