Title: A note on computational intelligence methods in biometrics

Authors: Qinghan Xiao

Addresses: Defence Research and Development Canada – Ottawa, 3701 Carling Avenue, Ottawa, Ontario, K1A 0Z4, Canada

Abstract: The research field of biometrics encompasses a wide range of disciplines such as signal processing, pattern recognition, classification, and decision making. Although various biometric algorithms are proposed, tested, reviewed and implemented every year, challenges still remain because of the existence of, among other issues, background noise, signal distortion, biometric template aging, and measurement environment variations. Computational intelligence, primarily based on artificial intelligence, artificial neural networks, fuzzy systems, and evolutionary computation, provides a powerful tool to deal with various important and pressing challenges in biometrics. This paper investigates state of the art computational intelligence techniques to deal with biometric problems that traditional techniques have found limited success.

Keywords: biometrics; computational intelligence; feature selection; matching; biometric fusion; artificial intelligence; artificial neural networks; ANNs; fuzzy logic; evolutionary computation.

DOI: 10.1504/IJBM.2012.046303

International Journal of Biometrics, 2012 Vol.4 No.2, pp.180 - 188

Published online: 25 Mar 2012 *

Full-text access for editors Access for subscribers Purchase this article Comment on this article