Biometric Recognition System Research Articles

Fusion of physiological measures for multimodal biometric systems

Physiological measures are widely studied from a medical point of view. Most applications lie in the field of diagnosis of heart attacks, as regards the ECG, or the detection of epileptic events, in the case of the EEG. In the last ten years, these signals are being investigated also from a biometric point of view, in order to exploit the discriminative capability provided by these measures in recognizing individuals. The present work proposes a multimodal biometric recognition system based on the fusion of the first lead (i) of the electrocardiogram (ECG) with six different bands of the electroencephalogram (EEG). The proposed approach is based on the extraction of fiducial features (peaks) from the ECG combined with spectrum features of the EEG. A dataset has been created, by composing the signals of two well-known databases. The results, reported by means of EER values, AUC values and ROC curves, show good recognition performances.

Multimodal biometrics recognition based on local fusion visual features and variational Bayesian extreme learning machine

Multimodal biometrics provides rich information in biometric recognition systems, thus a valid multimodal feature fusion framework and an efficient recognition algorithm are desirable for multimodal biometrics systems. In this paper, we design a multimodal fusion framework for face and fingerprint images using block based feature-image matrix, and extract a type of middle-layer semantic feature from local features—a local fusion visual feature, which has better characterization capabilities with lower dimension for multimodal biometrics. Furthermore, we create recognition utilizing the Variational Bayesian Extreme Learning Machine (VBELM), which has an obvious speed advantage by random input weights, and also has superior stability and generalization by adding a non-informative full Gaussian prior. This research enables multimodal biometrics recognition system to have a concentrated fusion feature description and great recognition performance. Experimental results show that the proposed multimodal biometrics recognition system has a higher testing accuracy in comparison to the traditional methods with higher efficiency and better stability.

Ekstraksi Ciri pada Citra Iris Menggunakan Gabor 2-D

Abstrak – Iris merupakan salah satu bagian tubuh manusia yang sering digunakan dalam sistem pengenalan biometrik karena tingkat perbedaannya yang tinggi. Ektraksi ciri merupakan salah satu tahapan yang dilalui dalam pengembangan sistem pengenalan biometrik iris. Tahap ini bertujuan untuk mengekstrak informasi dari citra iris yang telah disegmentasi sehingga dapat digunakan sebagai ciri unik dari iris bersangkutan. Pada paper ini tahap ekstraksi ciri dilakukan dengan menggunakan tapis Gabor 2-D. Tapis tersebut digunakan karena mampu menyediakan representasi gabungan yang optimal dari sinyal dalam domain spasial dan frekuensi. Hasil penerapan tapis Gabor 2-D didemodulasi dengan menggunakan quadrature Gabor 2-D untuk menghasilkan iris code yang dijadikan sebagai informasi pembeda (fitur ciri) iris. Hasil uji coba pada penelitian ini menghasilkan fitur iris terbaik ketika ukuran tapis yang digunakan adalah 33×33. Sudut orientasi yang digunakan untuk fitur real dan imaginary adalah -45º, 0º, 45º, dan 90º. DOI: 10.24843/MITE.1501.03

E-Passport Recognition System Based on ANNs

Nowadays it is more necessary to perform the identity check of passengers quickly and reliably to prevent unauthorized border crossing, and limit the use of forger passport. This paper concentrated on the design E-passport using two main technologies which are biometric and RFID technologies. Biometric features are used to identify passport holder and the RFID is used to store andtransmit these features as required. This paper proposes a new approach to design and implement a robust biometric recognition system that could be used in e-passport system to identify and recognize person that own the identical e-passport. The ANN is used for recognition persons in this proposed system which was able to recognize persons registered in database in rate up to 81% and the percentage of fail in recognition was 19%.

Iris Movements: The Best State to Dynamic Biometric Recognition Process

Substantially, biometric recognition systems that use the information on iris texture employ approximate models. They minimize the problems associated with iris dilation and contraction. This article demonstrates quantitatively that iris movements may lead to significant differences between the enrolment images from a database and the test image. In order to clarify this assumption two experiments were performed and the results indicate that the system performs better when the dilation rate is smaller. Unlike what was expected, images with greater dilation rate do not cause the worst system performance. Lastly, images with intermediate dilation rate brought the system to present the worst outcome.

Humans vs machines: the future of facial recognition

The dream of people discovering they have crime-fighting super-powers is the stuff of Hollywood movies like ‘Superman’ and the ‘X-Men’. But the startling fact about the individuals known as super recognisers is that their powers are real – they are literally living the dream. These people are supremely adept at facial recognition and their abilities have been harnessed by police to track the criminals behind the 2011 London riots and 2015 Cologne sex attacks. Yet in an era when humans are used to being bested by machines – witness the IBM super-computer that beat chess grandmaster Gary Kasparov and the Google program that recently humbled the world champion of the intuitive game ‘Go’ – the abilities of super recognisers have outshone current biometric facial recognition systems, raising questions about the technology. So what can developers of face recognition systems learn from them?

Effective Approches and Challenges Focused in IRIS Image Recognition – A Survey

One of the most well-known biometric identification iris recognition system is viewed as the most accurate and reliable biometric identification system. Over the past few years, although the False Rejection Rate and Equal Error Rate of iris images are increasing and False acceptance Rate of iris images are decreasing, the iris images causes few effects due to feature extraction approach. As a result feature extraction from the actual image has to be made to provide better authentication, which is a difficult task. Various techniques have been present in the literature on iris feature extraction which has its own limitations and advantages. The purpose of this article is to present a survey of the published literature which deals with the iris feature extraction, focusing on the relative merits and demerits of a variety of Iris recognition algorithms.

Face Recognition across Time Lapse Using Convolutional Neural Networks

Time lapse, characteristic of aging, is a complex process that affects the reliability and security of biometric face recognition systems. This paper reports the novel use and effectiveness of deep learning, in general, and convolutional neural networks (CNN), in particular, for automatic rather than hand-crafted feature extraction for robust face recognition across time lapse. A CNN architecture using the VGG-Face deep (neural network) learning is found to produce highly discriminative and interoperable features that are robust to aging variations even across a mix of biometric datasets. The features extracted show high inter-class and low intra-class variability leading to low generalization errors on aging datasets using ensembles of subspace discriminant classifiers. The classification results for the all-encompassing authentication methods proposed on the challenging FG-NET and MORPH datasets are competitive with state-of-the-art methods including commercial face recognition engines and are richer in functionality and interoperability than existing methods as it handles mixed biometric datasets, e.g., FG-NET and MORPH.

A Robust Multimodal Biometric System Integrating Iris, Face and Fingerprint using Multiple SVMs

This paper presents a robust multimodal biometric recognition system integrating iris, face and fingerprint based on match score level fusion using multiple support vector machines (SVMs). Here, multiple support vector machines are applied in parallel fashion to overcome the problem of missing biometric traits. It considers every possible combination of all the three biometric traits (iris, face and fingerprint) individually. Each possible combination of biometric traits has a separate SVM to combine the available match scores to generate the final decision. Existing multimodal biometric recognition systems are based on the assumption that the set of biometric traits to be integrated is always present as a whole at the time of authentication. But sometimes it is not possible due to some unavoidable circumstances (e.g. injury may be caused, person may be under some medical treatment, corresponding trait may be missing etc.). The performance of the proposed system is evaluated on a public dataset demonstrating its recognition accuracy regarding FAR (False Accept Rate) and FRR (False Reject Rate). Keywords: support vector machine (SVM), score level fusion, iris recognition, face recognition, fingerprint recognition, receiver operating characteristic (ROC) curve.

A Novel Approach to Brain Biometric User Recognition

Advances in the field of information technology makes information security is a major part of it. To deal with security, Recognition plays an important role. Now a day the numbers of organizations are using automated person recognition systems to improve customer satisfaction, to enhance operating efficiency and to secure critical resources. The reliable biometric recognition systems to either confirm or determine the individual's identity to servicing their requests. This paper gives a novel approach on user recognition using EEG signals of Brain. EEG is sensitive to the electrical field generated by the electric currents induced in the brain. EEG recordings are acquired with portable and relatively inexpensive devices when compared to the other brain imaging technique like fMRI. The Emotiv EPOC EEG neuro headset has 14 saline electrodes with two reference sensors is used for the acquisition of brainwaves.

On the Permanence of EEG Signals for Biometric Recognition

Brain signals have been investigated for more than a century in the medical field. However, despite the broad interest in clinical applications, their use as a biometric identifier has been only recently considered by the scientific community. In this paper, we focus on the permanence across time of brain signals, specifically of electroencephalographic (EEG) signals, issue of paramount importance for the deployment of brain-based biometric recognition systems in real life, not yet fully addressed. In particular, we speculate about the stability of EEG features by analyzing the recognition performance that can be achieved when comparing EEG signals acquired during different sessions. We carry out an extensive set of experimental tests, performed on several EEG-based biometric systems over a large database, comprising three recordings taken from 50 healthy subjects in resting state conditions, acquired in a time span of approximately one month and a half. The results confirm that a significant level of permanence can be guaranteed.

Non-stationary feature fusion of face and palmprint multimodal biometrics

Multimodal biometrics provides high performance in biometric recognition systems with respect to unibiometric systems as they offer a more universal approach, added security and better recognition accuracy. Moreover, data acquisition at the feature level brings out rich information from the traits, thus fusion of modalities at this level is desirable. In this paper we propose a novel fusion technique called non-stationary feature fusion where a new structure of interleaved matrix is constructed using local features extracted from two modalities i.e. face and palmprint images. A block based Discrete Cosine Transform (DCT) algorithm is used to construct a fused feature vector by extracting independent feature vectors from each spatial image. This fused feature vector contains nonlinear information that is used to train a Gaussian Mixture Models (GMM) based statistical model. The model provides accurate estimation of the class conditional probability density function of the fused feature vector. The proposed method produces recognition rates as high as 99.7% and 97% when tested on benchmark databases-ORL-PolyU and FERET-PolyU respectively. These rates are achieved using 23% low frequency DCT coefficients. The new technique is shown to outperform existing feature level fusion methods including methods based on matching and decision level fusion.

Multimodal Biometrics Recognition by using Modified Unconstrained Cohort Normalisation under Unconstrained Setting

Objective: The main intention of this research is to provide secured authentication on mobile devices with the use of multimodal based biometric authentication system under unconstrained settings. Methods: Method used in this research is pattern recognition algorithm namely modified unconstrained cohort normalisation (MUCN) is introduced into the score-level fusion process of multi-biometric system. The goal of proposed MUCN is normalizing the unconstraint modalities by correcting the misclassified scores occur in the UCN. Score normalization of multimodal biometric is enhanced and investigated for improving the accuracy performance of the multimodal biometric in an unconstraint setting. Results: The result of presented pattern recognition algorithm performs well in terms of recognition accuracy when compared to existing schemes.From the comparative evaluation on WVU multimodal data set, the proposed MUCN based Score level fusion achieves 89.2 % of overall recognition rate and out-performs existing state-of-art techniques. Conclusion: The present work demonstrates that the result obtained by MUCN can considerably improve the accuracy of fused biometrics. Thus it can be concluded that with respect to the obtained comparison results from the experiment, the proposed method provides highest recognization rate when compare with other conventional methods of biometric recognition system.

Towards the applicability of biometric wood log traceability using digital log end images

Log traceability in the timber based industries is a basic requirement to fulfil economical, social and legal requirements. This work introduces biometric log recognition using digital log end images and explores the robustness to a set of log end cross-section (CS) variations. In order to investigate longitudinal and surface CS variations three tree logs were sliced and captured in different sessions. A texture feature-based technique well known from fingerprint recognition is adopted to compute and match biometric templates of CS images captured from log ends. In the experimental evaluation insights and constraints on the general applicability and robustness of log end biometrics to identify logs in an industrial application are presented. Results for different identification performance scenarios indicate that the matching procedure which is based on annual ring pattern and shape information is very robust to log length cutting using different cutting tools. The findings of this study are a further step towards the development of a biometric log recognition system.

Research on Multimode Biometric Features Recognition System Adopting Neural Network

The small sample issue is a common problem in face recognition system, and multi-modal model has strong generalization ability to solve the problem of small sample, which has already become the most important area of research in pattern recognition, however, the low accuracy and efficiency of the model has become a major challenge. Based on this, this paper proposes a efficient multimode biometric face and fingerprint recognition system based on neural network, which provides more efficient identification though choosing a good feature extraction and recognition algorithms. The Adoption of biometric recognition to authenticate a person's identity has greatly improved operational efficiency and the recognition accuracy in comparison with adoption of password or passphrase. The feasibility and effectiveness of the method in this paper has been verified in multimode biometric system database.

Feature exploration for biometric recognition using millimetre wave body images

The use of millimetre wave images has been proposed recently in the biometric field to overcome certain limitations when using images acquired at visible frequencies. Furthermore, the security community has started using millimetre wave screening scanners in order to detect concealed objects. We believe we can exploit the use of these devices by incorporating biometric functionalities. This paper proposes a biometric recognition system based on the information of the silhouette of the human body, which may be seen as a type of soft biometric trait. To this aim, we report experimental results on the BIOGIGA database with four feature extraction approaches (contour coordinates, shape contexts, Fourier descriptors and landmarks) and three classification methods (Euclidean distance, dynamic time warping and support vector machines). The best configuration of 1.33 % EER is achieved when using contour coordinates with dynamic time warping.

Adversarial Biometric Recognition : A review on biometric system security from the adversarial machine-learning perspective

In this article, we review previous work on biometric security under a recent framework proposed in the field of adversarial machine learning. This allows us to highlight novel insights on the security of biometric systems when operating in the presence of intelligent and adaptive attackers that manipulate data to compromise normal system operation. We show how this framework enables the categorization of known and novel vulnerabilities of biometric recognition systems, along with the corresponding attacks, countermeasures, and defense mechanisms. We report two application examples, respectively showing how to fabricate a more effective face spoofing attack, and how to counter an attack that exploits an unknown vulnerability of an adaptive face-recognition system to compromise its face templates.

Robust multimodal face and fingerprint fusion in the presence of spoofing attacks

Anti-spoofing is attracting growing interest in biometrics, considering the variety of fake materials and new means to attack biometric recognition systems. New unseen materials continuously challenge state-of-the-art spoofing detectors, suggesting for additional systematic approaches to target anti-spoofing. By incorporating liveness scores into the biometric fusion process, recognition accuracy can be enhanced, but traditional sum-rule based fusion algorithms are known to be highly sensitive to single spoofed instances. This paper investigates 1-median filtering as a spoofing-resistant generalised alternative to the sum-rule targeting the problem of partial multibiometric spoofing where m out of n biometric sources to be combined are attacked. Augmenting previous work, this paper investigates the dynamic detection and rejection of liveness-recognition pair outliers for spoofed samples in true multi-modal configuration with its inherent challenge of normalisation. As a further contribution, bootstrap aggregating (bagging) classifiers for fingerprint spoof-detection algorithm is presented. Experiments on the latest face video databases (Idiap Replay-Attack Database and CASIA Face Anti-Spoofing Database) and fingerprint spoofing database (Fingerprint Liveness Detection Competition 2013) illustrate the efficiency of proposed techniques.

Extraction of Region of Interest (ROI) for Palm Print and Inner Knuckle Print

paper proposes an novel approach to extract the Region of interest (ROI) for palm print and inner knuckle print(IKP) from hand images captured from a digital camera. An new algorithm to detect candidate key points on hand region is described. Using the candidate key points a novel approach to locate the ROI of palm print and inner knuckle print is proposed. The proposed approach is evaluated using the database collected at our institute. The results obtained are promising and confirm the usefulness of proposed ROI extraction approach for developing hand based biometric recognition systems. Keywordssystems, ROI, Palm print, IKP.

Sampling variability in forensic likelihood-ratio computation: A simulation study

Recently, in the forensic biometric community, there is a growing interest to compute a metric called “likelihood-ratio” when a pair of biometric specimens is compared using a biometric recognition system. Generally, a biometric recognition system outputs a score and therefore a likelihood-ratio computation method is used to convert the score to a likelihood-ratio. The likelihood-ratio is the probability of the score given the hypothesis of the prosecution, Hp (the two biometric specimens arose from a same source), divided by the probability of the score given the hypothesis of the defense, Hd (the two biometric specimens arose from different sources). Given a set of training scores under Hp and a set of training scores under Hd, several methods exist to convert a score to a likelihood-ratio. In this work, we focus on the issue of sampling variability in the training sets and carry out a detailed empirical study to quantify its effect on commonly proposed likelihood-ratio computation methods. We study the effect of the sampling variability varying: 1) the shapes of the probability density functions which model the distributions of scores in the two training sets; 2) the sizes of the training sets and 3) the score for which a likelihood-ratio is computed. For this purpose, we introduce a simulation framework which can be used to study several properties of a likelihood-ratio computation method and to quantify the effect of sampling variability in the likelihood-ratio computation. It is empirically shown that the sampling variability can be considerable, particularly when the training sets are small. Furthermore, a given method of likelihood-ratio computation can behave very differently for different shapes of the probability density functions of the scores in the training sets and different scores for which likelihood-ratios are computed.