I. Introduction
Design of vision-based systems for traffic analysis is an important and challenging problem of great practical value. In the context of Intelligent Transportation Systems, the information added by image processing techniques is very useful and it comes at a very low computational load. Many works on ITS aim at helping traffic flow management by providing information on how many vehicles are in the scene. Moreover, incident detection, queue detection and measurement, intersection management, and many other applications could exploit such information provided by visual tasks. All the above mentioned ITS applications aim, as first step, at detecting vehicles in the scene in order to count them, build up a behaviour-based database to identify erroneous or unauthorized operations or simply track them [1].