Publication Type : Conference Paper
Publisher : International Review on Computers and Software
Source : International Review on Computers and Software, Volume 8, Number 8, p.1889-1900 (2013)
Url : http://www.scopus.com/inward/record.url?eid=2-s2.0-84885913225&partnerID=40&md5=c75b3d51d6ad60d7e9f33d4560c4cc1b
Campus : Coimbatore
School : School of Engineering
Department : Electronics and Communication
Verified : Yes
Year : 2013
Abstract : Biometric recognition has become a common and reliable way to authenticate the identity of a person. Multimodal biometric system utilizes two or more individual modalities so as to improve the recognition accuracy. The key to multimodal biometrics is the fusion of the various biometric data after feature extraction. In this paper, score level fusion technique for multi-modal biometric recognition using Artificial Bee Colony (ABC) based Neural Network (NN) is proposed. The technique consists of two phases namely feature extraction phase and score fusion phase. Features are extracted from the fingerprint, face and iris modalities in the feature extraction phase. Fusion of score value is carried out after obtaining the individual matching scores from the three modalities. Fusion of scores is based on neural network where, ABC algorithm is used as a training algorithm and based on the scores obtained from ABC-based neural network, the recognition is done. The implementation is done using MATLAB and the performance of the proposed technique is evaluated using FRR, FAR, accuracy and ROC curve. The proposed technique is compared with KNN technique and from the results we can see that our proposed technique has achieved better results by having lower FRR and FAR values and higher accuracy measure. © 2013 Praise Worthy Prize S.r.l. - All rights reserved.
Cite this Research Publication : Aravinth J. and Valarmathy, Sb, “Score-level fusion technique for multi-modal biometric recognition using ABC-based neural network”, International Review on Computers and Software, vol. 8, pp. 1889-1900, 2013.