Introduction

In recent years the importance of various authentication and security methods increased. Vehicle type recognition, if solved accurately, is beneficial e.g. for building authentications, police camera control systems on crossings to match the number-plate against the car make (car thefts, insurance frauds, etc.) The aim is to obtain reliable classification of a vehicle in the image from a multitude of possible classes using a limited number of examples. Although classification of road going vehicles has been a subject of interest in the past, e.g. traffic control systems and toll-levy automation, vehicle type recognition has not been considered at this level of accuracy. Instead, most systems either detect (classify vehicle and non-vehicle) or classify vehicles in broad categories such as cars, buses, heavy goods vehicles.