Score normalization in multimodal biometric systems

Jain, Anil K.; Nandakumar, Karthik; Ross, Arun

doi:10.1016/j.patcog.2005.01.012

Cited by 1,804 publications

(989 citation statements)

References 17 publications

(3 reference statements)

Supporting

Mentioning

925

Contrasting

Unclassified

Order By: Relevance

“…This system models the signature as a holistic multidimensional vector composed of the best performing 40-feature subset extracted in [20] from the total set of 100 global features described in [19]. In the present study, we used this 40-feature representation of the signatures normalizing each of them to the range [0,1] using tanh-estimators [25]. Finally, the similarity scores are computed using the Mahalanobis distance between the input vector and a statistical model of the attacked client estimated using a number of training signatures.…”

Section: On-line and Off-line Signature Recognition Systemsmentioning

confidence: 99%

On-line signature recognition through the combination of real dynamic data and synthetically generated static data

Galbally

Diaz

Ferrer

et al. 2015

Pattern Recognition

107

View full text Add to dashboard Cite

This is the author’s version of a work that was accepted for publication in Pattern Recognition . Changes resulting from the publishing process, such as peer review, editing, corrections, structural formatting, and other quality control mechanisms may not be reflected in this document. Changes may have been made to this work since it was submitted for publication. A definitive version was subsequently published in Pattern Recognition , 48, 9 (2005) DOI: 10.1016/j.patcog.2015.03.019On-line signature verification still remains a challenging task within biometrics. Due to their behavioral nature (opposed to anatomic biometric traits), signatures present a notable variability even between successive realizations. This leads to higher error rates than other largely used modalities such as iris or fingerprints and is one of the main reasons for the relatively slow deployment of this technology. As a step towards the improvement of signature recognition accuracy, the present paper explores and evaluates a novel approach that takes advantage of the performance boost that can be reached through the fusion of on-line and off-line signatures. In order to exploit the complementarity of the two modalities, we propose a method for the generation of enhanced synthetic static samples from on-line data. Such synthetic off-line signatures are used on a new on-line signature recognition architecture based on the combination of both types of data: real on-line samples and artificial off-line signatures synthesized from the real data. The new on-line recognition approach is evaluated on a public benchmark containing both real versions (on-line and off-line) of the exact same signatures. Different findings and conclusions are drawn regarding the discriminative power of on-line and off-line signatures and of their potential combination both in the random and skilled impostors scenarios.M. D.-C. is supported by a PhD fellowship from the ULPGC and M.G.-B. is supported by a FPU fellowship from the Spanish MECD. This work has been partially supported by projects: MCINN TEC2012-38630- C04-02, Bio-Shield (TEC2012-34881) from Spanish MINECO, BEAT (FP7-SEC-284989) from EU, CECABANK and Cátedra UAM-Telefónic

show abstract

Section: On-line and Off-line Signature Recognition Systemsmentioning

confidence: 99%

On-line signature recognition through the combination of real dynamic data and synthetically generated static data

Galbally

Diaz

Ferrer

et al. 2015

Pattern Recognition

107

View full text Add to dashboard Cite

show abstract

“…This basic fusion method consists of averaging the matching scores provided by the different matchers. Under some mild statistical assumptions [20,21] and with the proper matching score normalization [22], this simple method is demonstrated to give good results for the biometric authentication problem. This fact is corroborated in a number of studies [21,23].…”

Section: Quality-based Score Fusionmentioning

confidence: 99%

Incorporating Image Quality in Multi-algorithm Fingerprint Verification

Fiérrez-Aguilar

Chen

Ortega‐Garcia

et al. 2005

Advances in Biometrics

View full text Add to dashboard Cite

Esta es la versión de autor de la comunicación de congreso publicada en: This is an author produced version of a paper published in: Abstract. The effect of image quality on the performance of fingerprint verification is studied. In particular, we investigate the performance of two fingerprint matchers based on minutiae and ridge information as well as their score-level combination under varying fingerprint image quality. The ridge-based system is found to be more robust to image quality degradation than the minutiae-based system. We exploit this fact by introducing an adaptive score fusion scheme based on automatic quality estimation in the spatial frequency domain. The proposed scheme leads to enhanced performance over a wide range of fingerprint image quality.

show abstract

“…We present two approaches for combining evidence based on generalized densities: (i) the product rule, which assumes independence between the individual modalities, and (ii) copula models, which parametrically model the dependence between the matching scores of multiple modalities. Our proposed method bypasses the need for score normalization and selection of optimal weights for the score combination on a case-by-case basis [3,9,10], and therefore, is a more principled approach with performance comparable to the commonly used fusion methods. Experiments have shown that our method achieves consistently high performance over the MSU and NIST multimodal databases.…”

Section: Introductionmentioning

confidence: 99%

A Principled Approach to Score Level Fusion in Multimodal Biometric Systems

Dass

Nandakumar

Jain

2005

Lecture Notes in Computer Science

Self Cite

136

View full text Add to dashboard Cite

Abstract. A multimodal biometric system integrates information from multiple biometric sources to compensate for the limitations in performance of each individual biometric system. We propose an optimal framework for combining the matching scores from multiple modalities using the likelihood ratio statistic computed using the generalized densities estimated from the genuine and impostor matching scores. The motivation for using generalized densities is that some parts of the score distributions can be discrete in nature; thus, estimating the distribution using continuous densities may be inappropriate. We present two approaches for combining evidence based on generalized densities: (i) the product rule, which assumes independence between the individual modalities, and (ii) copula models, which consider the dependence between the matching scores of multiple modalities. Experiments on the MSU and NIST multimodal databases show that both fusion rules achieve consistently high performance without adjusting for optimal weights for fusion and score normalization on a case-by-case basis.

show abstract

Score normalization in multimodal biometric systems

Cited by 1,804 publications

References 17 publications

On-line signature recognition through the combination of real dynamic data and synthetically generated static data

On-line signature recognition through the combination of real dynamic data and synthetically generated static data

Incorporating Image Quality in Multi-algorithm Fingerprint Verification

A Principled Approach to Score Level Fusion in Multimodal Biometric Systems

Contact Info

Product

Resources

About