Performance evaluation of score level fusion in multimodal biometric systems

He, Mingyi; Horng, Shi–Jinn; Fan, Pingzhi; Run, Ray-Shine; Chen, Rong-Jian; Lai, Jui-Lin; Khan, Muhammad Khurram; Sentosa, Kevin Octavius

doi:10.1016/j.patcog.2009.11.018

Cited by 211 publications

(119 citation statements)

References 27 publications

Supporting

Mentioning

118

Contrasting

Unclassified

Order By: Relevance

“…Multiple biometrics is an effective way to improve a performance of recognition system, which includes different sources [22] , whereas these sources may come from (a) multiple biometrics traits, such as iris and palmprint [23] , finger vein, fingerprint and face [24] (and so on), or ( [25] . However, although the fused feature vector has richer source of information, features from these modalities may not be compatible.…”

Section: ⅳ Matching Score-level Fusionmentioning

confidence: 99%

Finger Vein Recognition based on Matching Score-Level Fusion of Gabor Features

Yu¹,

Yoon²,

Park³

2013

The Journal of Korean Institute of Communications and Informati

View full text Add to dashboard Cite

Most methods for fusion-based finger vein recognition were to fuse different features or matching scores from more than one trait to improve performance. To overcome the shortcomings of "the curse of dimensionality" and additional running time in feature extraction, in this paper, we propose a finger vein recognition technology based on matching score-level fusion of a single trait. To enhance the quality of finger vein image, the contrast-limited adaptive histogram equalization (CLAHE) method is utilized and it improves the local contrast of normalized image after ROI detection. Gabor features are then extracted from eight channels based on a bank of Gabor filters. Instead of using the features for the recognition directly, we analyze the contributions of Gabor feature from each channel and apply a weighted matching score-level fusion rule to get the final matching score, which will be used for the last recognition. Experimental results demonstrate the CLAHE method is effective to enhance the finger vein image quality and the proposed matching score-level fusion shows better recognition performance.

show abstract

Section: ⅳ Matching Score-level Fusionmentioning

confidence: 99%

Finger Vein Recognition based on Matching Score-Level Fusion of Gabor Features

Yu¹,

Yoon²,

Park³

2013

The Journal of Korean Institute of Communications and Informati

View full text Add to dashboard Cite

show abstract

“…This database contains true similarity scores of face and fingerprints which belong to one person. This dataset is suited to the study of score-level fusion-based multimodal, multi-algorithmic and multi-sample biometrics (He et al, 2010;Nandakumar, Chen, Dass, & Jain, 2008). Two commercial face matchers, labeled with C and G, have generated similarity scores based on a frontal face comparison and a public fingerprint matcher has generated one left and one right index matching score, denoted by LI and RI respectively.…”

Section: Matching Scores Databasementioning

confidence: 99%

“…Some of these studies (Fox, Gross, Cohn, & Reilly, 2007;He et al, 2010;Monwar & Gavrilova, 2008;Wang & Han, 2008) involve robustness and security of authentication systems. In most studies, the researchers have considered the security from the viewpoint of attempt to decrease both FAR and FRR error rates by applying various identification methods and increasing the number of traits (from 2 to 3); without any major study on importance of robustness for a biometric system against attacks.…”

Section: Introductionmentioning

confidence: 99%

Robustness of Multi Biometric Authentication Systems against Spoofing

Hariri¹,

Shokouhi²

2011

CIS

View full text Add to dashboard Cite

Nowadays biometric authentication systems have been more developed, especially in secure and financial systems; so cracking a biometric authentication system is now a growing concern. But their security has not received enough attention. Imitating a biometric trait of a genuine user to deceive a system, spoofing, is the most important attacking method. Multi biometric systems have been developed to overcome some weaknesses of single biometric systems because the forger needs to imitate more than one trait. No research has further investigated the vulnerability of multimodal systems against spoof attack. We empirically examine the robustness of five fixed rules combining similarity scores of face and fingerprint traits in a bimodal system. By producing different spoof scores, the robustness of fixed combination rules is examined against various possibilities of spoofing. Robustness of a multi biometric system depends on the combination rule, the spoof trait and the intensity of spoofing. Min rule shows the most robustness when face is spoofed especially in very secure systems but when the fingerprint is faked the max rule shows the least vulnerability against possibilities of spoofing.

show abstract

“…This has motivated researchers in multi-biometric systems [3] to consolidate the evidence obtained from different sources. By using multi-biometric approaches, better performance requirement can be achieved as reported by several researches, for instances in [3]- [8].…”

Section: Introductionmentioning

confidence: 96%

Fusion of Face and Fingerprint for Robust Personal Verification System

Hassan¹,

Ramli²,

Suandi³

2014

IJMLC

View full text Add to dashboard Cite

Abstract-Personal verification system that uses a single biometric trait often faces numerous limitations such as noisy sensor data, non-universality, non-distinctiveness and spoof attack. These limitations can be overcome by multimodal biometric systems that consolidate the evidence presented by multiple biometric sources and typically has better recognition performance compared to systems based on a single biometric modality. This study proposes fusion of face and fingerprint for robust recognition system. The integration is performed at the matching score level. The matching tasks for both modalities are carried out by using support vector machines (SVM) as the classifier. Experiments on face expression and fingerprint database show that the performances of multimodal biometric system provide better recognition compared to single biometric modality. Based on the fusion techniques evaluated, trait-specific weight was found to be highly effective than the sum rule-based fusion. Equal error rate (EER) percentage for face-only and fingerprint-only systems are 2.50% and 5.56%, respectively, while the EER for system using sum rule-based fusion and system using trait-specific weights are 0.833% and 0.340%, respectively.Index Terms-Multi-modal, sum-rule and trait-specific, face and fingerprint biometrics.

show abstract

Performance evaluation of score level fusion in multimodal biometric systems

Cited by 211 publications

References 27 publications

Finger Vein Recognition based on Matching Score-Level Fusion of Gabor Features

Finger Vein Recognition based on Matching Score-Level Fusion of Gabor Features

Robustness of Multi Biometric Authentication Systems against Spoofing

Fusion of Face and Fingerprint for Robust Personal Verification System

Contact Info

Product

Resources

About