Restoring algorithm for traffic data based on self-adaptive generation of area geometry

Abstract

Intelligent Transportation System (ITS) is provided with basic data support and continuous motive force by traffic information. So the quality of the raw traffic data detected by traffic sensors will directly affect the follow-up benefits of the entire system. Traditional restoration processing method, such as algorithms based on historical trend data and linear interpolation, faded in shortage for data processing, so that the true and implicit orderliness in the traffic flow data can not be reflected. In order to improve the accuracy of raw traffic data, a traffic data restoring algorithm based on self-adaptive generation of area geometry is proposed in this paper, which can judge and restore the incomplete traffic data after validity test. Through the validation using the Beijing actual traffic data, it is proved that this algorithm is precise and reliable compared with several common data restoring algorithms.