Title: On selecting Gabor features for biometric authentication

Authors: Loris Nanni, Alessandra Lumini

Addresses: DEIS, IEIIT – CNR, Universita di Bologna, Viale Risorgimento 2, 40136 Bologna, Italy. ' DEIS, IEIIT – CNR, Universita di Bologna, Viale Risorgimento 2, 40136 Bologna, Italy

Abstract: We describe a Gabor feature selection technique that allows us to develop a fast and robust Gabor-feature based biometric system. Existing Gabor based methods use a huge number of Gabor features to represent the patterns. Our experiments on different biometric characteristics show that by using only few (∼10) Gabor features, it is possible to achieve a very low Equal Error Rate (EER). Only a low number of Gabor filters and sub-images, selected by running the Sequential Forward Floating Selection (SFFS), are exploited in the fusion step. The system has been tested on two biometric traits: ear authentication and finger authentication.

Keywords: Gabor features; feature selection; ear authentication; finger authentication; biometrics; equal error rate; automated identification.

DOI: 10.1504/IJCAT.2009.024592

International Journal of Computer Applications in Technology, 2009 Vol.35 No.1, pp.23 - 28

Published online: 10 Apr 2009 *

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