Vehicle OD Estimation Model Based on License Plate Recognition Data

Abstract

OD matrix of motor vehicles is the core data for urban road traffic network analysis. Based on the analysis of vehicle license plate recognition and detection data, the turning flow at Road intersection and some measured vehicle OD information in the whole network are used. The generalized least squares model is used to establish the OD estimation model of road network which integrates part of vehicle OD information. The OD historical value and distribution matrix in the model are applied to the real part. The OD information of the sub-motor vehicle is deduced. At the same time, in order to verify the impact of the proportion of detection data, the model is validated by using two different scale real road network detection data of the same city and combining with S-Paramics simulation platform. The results show that different detection proportions have a significant impact on OD estimation results, and the accuracy of OD estimation can be improved by using steering flow and part of vehicle OD information in high detection proportions.