Inferring and Analysis Drivers Violation Behavior Through Trajectory

Abstract

In this paper, we present an algorithm for inferring violation movements and categorizing levels of driving behavior. With this algorithm we extract the speeding and retrograde behavior from the real trajectories datasets of Xinjiang, analyze the changing regulation of six streets in the working day and day off on the overall and explore the driving characteristics which is very dangerous(level 4). The results of this study can not only be used for early warning of drivers violations, but also provide the data support and decision basis for the traffic management department to master the situation of traffic violations and formulate the traffic management counter measures.