%0 Journal Article
%T Four Machine Learning Algorithms for Biometrics Fusion: A Comparative Study
%A I. G. Damousis
%A S. Argyropoulos
%J Applied Computational Intelligence and Soft Computing
%D 2012
%I Hindawi Publishing Corporation
%R 10.1155/2012/242401
%X We examine the efficiency of four machine learning algorithms for the fusion of several biometrics modalities to create a multimodal biometrics security system. The algorithms examined are Gaussian Mixture Models (GMMs), Artificial Neural Networks (ANNs), Fuzzy Expert Systems (FESs), and Support Vector Machines (SVMs). The fusion of biometrics leads to security systems that exhibit higher recognition rates and lower false alarms compared to unimodal biometric security systems. Supervised learning was carried out using a number of patterns from a well-known benchmark biometrics database, and the validation/testing took place with patterns from the same database which were not included in the training dataset. The comparison of the algorithms reveals that the biometrics fusion system is superior to the original unimodal systems and also other fusion schemes found in the literature. 1. Introduction Identity verification has many real-life applications such as access control and economic or other transactions. Biometrics measure the unique physical or behavioural characteristics of an individual as a means to recognize or authenticate their identity. Common physical biometrics include fingerprints, hand or palm geometry, and retina, iris, or facial characteristics. On the other hand, behavioural characteristics include signature, voice (which also has a physical component), keystroke pattern, and gait. Although some technologies have gained more acceptance than others, it is beyond doubt that the field of access control and biometrics as a whole shows great potential for use in end user segments, such as airports, stadiums, defence installations, but also the industry and corporate workplaces where security and privacy are required. Biometrics may be used for identity establishment. A new measurement that purports to belong to a particular entity is compared against the data stored in relation to that entity. If the measurements match, the assertion that the person is whom they say they are is regarded as being authenticated. Some building access schemes work in this way, with the system comparing the new measure against the company¡¯s employee database. Also authentication of the identity of a person is frequently used in order to grand access to premises or data. Since authentication takes place instantaneously and usually only once, identity fraud is possible. An attacker can bypass the biometrics authentication system and continue undisturbed. A cracked or stolen biometric system presents a difficult problem. Unlike passwords or smart cards, which can be
%U http://www.hindawi.com/journals/acisc/2012/242401/