Selected Aspects of Traffic Signs Recognition: Visual versus RFID Approach

Abstract

Many scientific methods of traffic signs recognition involving digital image analysis have been proposed. Most of them are appearance-based approaches, employing template matching. In most cases they work on color images (or videos) and deal with all types of signs, regarding their shape and color. On the other hand, commercial systems, installed in higher-class cars, detect only the round speed limit signs and overtaking restrictions found all across Europe. The main disadvantage of visual recognition of traffic signs is associated with difficult conditions of image acquisition and hence problems with noise, blurring, scale and orientation changes should be solved. In the paper we present a classification of signs visual recognition methods and discuss their advantages and disadvantages. We compare them with an RFID approach. © Springer-Verlag Berlin Heidelberg 2013.