Elsevier

Neurocomputing

Volume 69, Issues 13–15, August 2006, Pages 1706-1710
Neurocomputing

Letters
An advanced multi-modal method for human authentication featuring biometrics data and tokenised random numbers

https://doi.org/10.1016/j.neucom.2006.01.010Get rights and content

Abstract

In this work, we propose a multi-modal method that combines the scores of selected fingerprint matchers with the scores obtained by a Face Authenticator where the facial features are combined with pseudo-random numbers. We propose a novel method to combine the scores of fingerprint matchers based on random subspace Ensemble and we test the method on the systems submitted to FVC2004. Moreover, we show that methods based on tokenised pseudo-random numbers and user specific biometric features are highly dependent upon a parameter, the hashing threshold; we demonstrate that using an ensemble of classifiers it is possible to solve this problem leading to a considerable performance improvement. Finally, we study the fusion among the scores obtained by a Face Authenticator (where the face features are combined with pseudo-random numbers) and the scores of the systems submitted to FVC2004.

Introduction

The increasing interest in a wide variety of practical applications for automatic personal identification and authentication has resulted in the popularity of biometric recognition systems [8]. As a consequence, recent efforts have been conducted in order to establish common evaluation scenarios enabling a fair comparison between competing systems. For instance, in the field of fingerprint recognition, a series of International Fingerprint Verification Competitions (FVC) [7] have received great attention both from the academy and the industry. Denial of access in biometric systems greatly impacts on the usability of the system by failing to identify a genuine user. Multimodal biometrics can reduce the probability of denial of access without sacrificing the false acceptation performance, the key is the combination of the various biometric characteristics at the feature extraction, match score, or decision level [8]. In [4], the authors, in order to solve the problem of high false rejection, proposed a novel two-factor authenticator based on iterated inner products between tokenised pseudo-random numbers and the user fingerprint feature; in this way a set of user compact codes can be produced which is named “BioHashing”. The possible drawback of BioHashing is the low performance when an “impostor” B steals the pseudo-random numbers (hash key) of A and tries to authenticate as A. When this problem occurs, the performance of BioHashing can be lower than that obtained using only the biometric data. In [6], the authors proposed a multimodal system based on the fusion between “BioHashed” face features (face features combined with pseudo-random numbers) and the scores obtained by some of the fingerprint verification methods submitted to FVC2004. They showed that the fusion permits to obtain good performance (similar to that obtained by the standard fusion between face and fingerprint) also when an “impostor” steals an hash key.

In this paper, we improve that result by proposing to use a random subspace (RS) method [3] to combine the scores of fingerprint matchers. We show that a fusion of classifiers based on RS permits to obtain an equal error rate (EER) lower than that obtained by a “stand-alone” classifier. Moreover, we show that, differently from what is assumed by previous works, the performance of a BioHashing system is dependent upon a parameter [4], [6] (see Section 2 for details), and we give a method for making it independent from that. We suggest to vary the value of this parameter and to train a classifier for each value and finally to combine these classifiers: experimental results prove that the suggested method obtains an EER lower than the best EER obtained by tuning the parameter. In the following sections we briefly review the BioHashing method (Section 2), we detail our multi-modal method for human authentication (Section 3), then we present and discuss some experimental results (Section 4) and we draw our conclusions (Section 5).

Section snippets

Biohashing

The biometric vector data xRM is reducing down to a set of single bits b{0,1}m with m the length of the bit string (in this paper we fix m to 200), via tokenised pseudo random patterns riRM. In our implementation we use as pseudo random number generator the Blum–Blum–Shub method [1]. In the following the algorithm for the creation of the vector b is detailed:

  • (1)

    generate a set of pseudo random vectors {riRM,i=1,,m},

  • (2)

    apply the Gram-Schmidt process to transform the basis {ri} into an orthonormal

The proposed method

A popular approach to combining multiple classifiers in biometric recognition is to treat the combination stage as a second-level pattern recognition problem on the matching scores that are to be fused [2], and then use standard learning paradigms in order to obtain combining functions. The similarity score output of each system is seen as a different feature, and the two classes correspond to impostor and genuine attempts, respectively. Comparative studies in this field [2] show that the

Experiments and discussion

Details of the FVC2004 competition and results were presented in [7]. Two different sub-competitions (open and light) were organized. The light category was intended for algorithms with restricted computing and memory usage; 26 algorithms participated in this case. 41 algorithms participated in the open category. In this work, we focus on combining the algorithms competing in the open category. Data for the competition consists of four different databases. It is worth noting, the image quality

Conclusions

In this paper, we proposed the fusion among the score obtained by a FA (where the face features are combined with pseudo-random numbers) and the scores of some selected systems submitted to FVC2004. We have shown that an ensemble of classifiers can be used to combine the scores of fingerprint matchers and, for boosting the performance, of a FA where the face features are combined with pseudo-random numbers. Our approach obtains an amazing near zero-EER on FVC2004, this means that our multimodal

Acknowledgements

This work has been supported by Italian PRIN prot. 2004098034 and by European Commission IST-2002-507634 Biosecure NoE projects.

Loris Nanni is a Ph.D. Candidate in Computer Engineering at the University of Bologna, Italy. He received his Master Degree cum laude in 2002 from the University of Bologna. In 2002, he starts his Ph.D. in Computer Engineering at DEIS, University of Bologna. His research interests include pattern recognition, and biometric systems (fingerprint classification and recognition, signature verification, face recognition).

References (12)

There are more references available in the full text version of this article.

Cited by (18)

  • Enhanced computer vision applications with blockchain: A review of applications and opportunities

    2023, Journal of King Saud University - Computer and Information Sciences
  • Soft biometrics in conjunction with optics based biohashing

    2011, Optics Communications
    Citation Excerpt :

    An improved biohashing technique has also been proposed which shows the good performance when the impostor steals the hash key [14]. Multimodal biometric system based on the use of a number of biometric traits simultaneously has also been proposed in order to improve the performance of the biometrics system [17–22]. Support vector machines (SVM), weighted SUM of scores, multilayer perceptron (MLP) as well as logical OR and AND are the methods employed for fusion of different biometrics [19,21–23].

  • Optics based biohashing using joint transform correlator

    2010, Optics Communications
    Citation Excerpt :

    Biohash code generation has received much attention these days in biometric recognition [17–23].

View all citing articles on Scopus

Loris Nanni is a Ph.D. Candidate in Computer Engineering at the University of Bologna, Italy. He received his Master Degree cum laude in 2002 from the University of Bologna. In 2002, he starts his Ph.D. in Computer Engineering at DEIS, University of Bologna. His research interests include pattern recognition, and biometric systems (fingerprint classification and recognition, signature verification, face recognition).

Alessandra Lumini received a degree in Computer Science from the University of Bologna, Italy, on March 26, 1996. In 1998, she started her Ph.D. studies at DEIS—University of Bologna and in 2001 she received her Ph.D. degree for her work on “Image Databases”. Now she is an Associate Researcher at University of Bologna. She is a member of the BIAS Research Group at the department of Computer Science of the University of Bologna (Cesena). She is interested in Biometric Systems (particularly Fingerprint Classification), Multidimensional Data Structures, Digital Image Watermarking and Image Generation.

View full text