554 Ensemble of ANN for Traffc Sign Recognition M. Paz Sesmero Lorente Universidad Carlos III de Madrid, Spain Juan Manuel Alonso-Weber Universidad Carlos III de Madrid, Spain Germán Gutiérrez Sánchez Universidad Carlos III de Madrid, Spain Agapito Ledezma Espino Universidad Carlos III de Madrid, Spain Araceli Sanchis de Miguel Universidad Carlos III de Madrid, Spain Copyright © 2009, IGI Global, distributing in print or electronic forms without written permission of IGI Global is prohibited. INTRODUCTION “Machine Learning (ML) is the subfeld of Artifcial Intelligence conceived with the bold objective to de- velop computational methods that would implement various forms of learning, in particular mechanisms capable of inducing knowledge form examples or data” (Kubat, Bratko & Michalski, 1998, p. 3). The simplest and best-understood ML task is known as supervised learning. In supervised learning, each example consists of a vector of features (x) and a class (y). The goal of the learning algorithm is, given a set of examples and their classes, fnd a function, f, that can be applied to assign the correct class to new examples. When the function f takes values from a discrete set of classes {C 1 , .…., C K ,}, f is called a classifer (Diet- terich, 2002). In the last decades it has been proved that learning tasks in which the unknown function f takes more than two values (multi-class learning problems) the better approach is to decompose the problem into multiple two-class classifcation problems (Ou & Murphey, 2007) (Dietterich, & Bakiri, 1995) (Massulli & Val- entini, 2000). This article describes the implementation of a system whose main task is to classify prohibition road signs into several categories. In order to reduce the learning problem complexity and to improve the classifcation performance, the system is composed by a collection (ensemble) of independent binary classifers. In the proposed approach, each binary classifer is a single- output neural network (NN) trained to distinguish a particular road sign kind from the others. The proposed system is a part of a Driver Support System (DSS) supported by the Spanish Government under project TRA2004-07441-C03-C02. For this reason, one of the main system requirements is that it should be implemented in hardware in order to use it aboard a vehicle for real time categorization. In order to fulfll this constraint, a reduction in the number of features that describe the instances must be performed. As consequence if we have k generic road sign types we will use k binary NN and k feature selection process will be executed. BACKGROUND It is known that road signs carry essential information for safe driving. Among other things, they permit or prohibit certain maneuvers, warn about risk factors, set speed limits and provide information about directions, destinations, etc. Therefore, road sign recognition is an essential task for the development of an autonomous Driver Support System. In spite of the increasing interest in the last years, traffc sign recognition is one of the less studied subjects in the feld of Intelligent Transport Systems. Approaches in this area have been mainly focused on the resolution of other problems, such as road border detection (Dickmanns & Zapp, 1986) (Pomerlau & Jochem, 1996) or the recognition of obstacles in the