Figure 2 - uploaded by Zahid Akhtar
Content may be subject to copyright.
6: A multimodal biometric system made up of a fingerprint and a face sensor, whose match scores are combined through a fusion rule.
Source publication
Similar publications
Citations
... One approach to increase the robustness against spoofing attacks is to use multimodal biometrics. However, multimodal systems often consider the different modalities independently and, since they often focus on the trait with the least distortions, they are vulnerable against spoofing attacks imitating only one trait successfully [3]. ...
Acoustic speaker recognition systems are very vulnerable to spoofing attacks via replayed or synthesized utterances. One possible countermeasure is audio-visual speaker recognition. Nevertheless, the addition of the visual stream alone does not prevent spoofing attacks completely and only provides further information to assess the authenticity of the utterance. Many systems consider audio and video modalities independently and can easily be spoofed by imitating only a single modality or by a bimodal replay attack with a victim’s photograph or video. Therefore, we propose the simultaneous verification of the data synchronicity and the transcription in a challenge-response setup. We use coupled hidden Markov models (CHMMs) for a text-dependent spoofing detection and introduce new features that provide information about the transcriptions of the utter- ance and the synchronicity of both streams. We evaluate the features for various spoofing scenarios and show that the com- bination of the features leads to a more robust recognition. Ad- ditionally, by evaluating the data on unseen speakers, we show the spoofing detection to be applicable in speaker-independent use-cases.
... Multimodal biometrics systems have proven to solve some problems associated with unimodal systems. Unimodal systems suffer from problems of intra-class distinctions, noise, inflexibility, non-universality, spoof attacks and high error rates[23]. Multimodal systems are able to protect itself from these complications. Intra-class distinctions basically mean that data is spread over a large plane making it difficult to classify data. ...
Authentication is the process of validating the identity of a person based on certain input that the person provides. Authentication has become a major topic of research due to the increasing number of attacks on computer networks around the globe. This review paper focuses on multimodal biometric authentication systems in use today. The aim is to elicit the best combination of authentication factors for multimodal use. We study the strengths and weakness of selected biometric mechanisms and recommend novel solutions to include in multimodal biometric systems to improve on the current biometric drawbacks. We believe this paper will provide security researchers some useful insight whilst designing better biometric systems.
... The biometric based verification system is essentially a pattern-recognition system that recognizes a person based on a feature vector extracted from physical or behavioral traits. This type of systems has evolved to play a critical role in personal, national and global security [1] , which includes a variety of applications that require reliable verification techniques, such as passports, border control, and used in computer systems, cellular phones, medical records management, banking. The systems are also used in important and sensitive government departments , which include the traffic offices. ...
... The extracted features are averaged and this becomes the training set which is otherwise called the system database that is stored for future matching process for the purpose of identification. Second step is the recognition stage in which the input is compared with the database and at last the ranking process is done to obtain the results [10][11][12][13]. ...
In the real world applications, wireless networks are an integral part of day-to-day life for many people, with businesses and home users relying on them for connectivity and communication. This paper examines the problems relating to the topic of wireless security and the background literature. The biometric systems often face limitations because of sensitivity to noise, intra class invariability, data quality, and other factors. Improving the performance of individual matchers in the aforementioned situation may not be effective. Multi biometric systems are used to overcome this problem by providing multiple pieces of evidence of the same identity. This system provides effective fusion scheme that combines information presented by the multiple domain experts based on the Rank level fusion integration method, thereby increasing the efficiency of the system which is not possible by the unimodal biometric system. The proposed multimodal biometric system has a number of unique qualities, starting from utilizing principal component analysis and fisher's linear discriminant methods for individual matchers authentication and the novel rank level fusion method is used in order to consolidate the results obtained from different biometric matchers. The ranks of the individual matchers are combined using highest rank, Borda count, and logistic regression method. From the results it can be concluded that the overall performances of the wireless security based multi biometric systems are improved even in the presence of quality of data.
... Fingerprints are used for authentication in various security applications based on their distinctive features [1]. Despite the recent progress, fingerprint recognition systems are vulnerable to spoof attack, which consists in submitting to the system an artefact fingerprint [2, 3]. Spoof attacks are a major issue for companies willing to market biometric based identity management solutions. ...
Fingerprint recognition systems are vulnerable to spoof attacks, which consist in presenting forged fingerprints to the sensor. Typical anti-spoofing mechanism is fingerprint liveness detection. Existing liveness detection methods are still not robust to spoofing materials, datasets and sensor variations. In particular, the performance of a liveness detection algorithm remarkably drops upon encountering spoof fabrication materials that were not used during the training stage. Likewise, a quintessential liveness detection method needs to be adapted and retrained to new spoofing materials, datasets and each sensor used for acquiring the fingerprints. In this paper, we propose a framework that first performs correlation mapping between live and spoof fingerprints and then uses a discriminative-generative classification scheme for spoof detection. Partial Least Squares (PLS) is utilized to learn the correlations. While, support vector machine (SVM) is combined with three generative classifiers, namely Gaussian Mixture Model, Gaussian Copula, and Quadratic Discriminant Analysis, for final classification. Experiments on the publicly available LivDet2011 and LivDet2013 datasets, show that the proposed method outperforms the existing methods alongside cross-spoof material and cross-sensor techniques.
... Vulnerability of biometric systems to spoof attacks is well known [1, 2, 3, 4, 5, 6] . A spoof attack occurs when an adversary mimics the biometric trait of another individual in order to circumvent the system. ...
... To minimize the impact of fabrication materials, denoised version of 1000 live as well as 1000 fake fingerprint images (via linear as well as wavelet based non-linear denoising) were used. We used a gaussian filter (window size = [5, 5] pixels) for linear filtering and symlet-based wavelet (with thresholding type set to hard, decomposition level set to 5 and threshold set to 7 (we tested threshold values from 5 to 15 in increments of 2)) 3 for non-linear denoising. In the training stage, rotation invariant uniform LBP patterns were extracted from fingerprint images denoised using gaussian and symlet-based wavelet denoising, separately, and two SVMs were trained. ...
... To minimize the impact of fabrication materials, denoised version of 1000 live as well as 1000 fake fingerprint images (via linear as well as wavelet based non-linear denoising) were used. We used a gaussian filter (window size = [5, 5] pixels) for linear filtering and symlet-based wavelet (with thresholding type set to hard, decomposition level set to 5 and threshold set to 7 (we tested threshold values from 5 to 15 in increments of 2)) 3 for non-linear denoising. In the training stage, rotation invariant uniform LBP patterns were extracted from fingerprint images denoised using gaussian and symlet-based wavelet denoising, separately, and two SVMs were trained. ...
Fingerprint liveness detection algorithms have been used to disambiguate live fingerprint samples from spoof (fake) fingerprints fabricated using materials such as latex, gelatine, etc. Most liveness detection algorithms are learning-based and dependent on the fabrication materials used to generate spoofs during the training stage. Consequently, the performance of a liveness detector is significantly degraded upon encountering fabrication materials that were not used during the training stage. The aim of this work is to design a simple pre-processing scheme that can improve the interoperability of liveness detectors across different fabrication materials - including those not observed during the training stage. Such a generalization ability is desirable in liveness detectors. Experiments on the LivDet 2011 fake fingerprint dataset suggest that (a) different fabrication materials when used in the training stage impart different degrees of generalization ability to the liveness detector and (b) the proposed pre-processing scheme improves generalization performance by upto 44%.
... bile devices, vulnerabilities to spoofing attacks are mainly overlooked. Spoofing is the process by which a fraudulent user can subvert or attack a biometric system by masquerading as registered user and thereby gaining illegitimate access and advantages [1, 2]. As illustrated in Fig. 1, face, iris and fingerprint images captured from spoofing attacks can look very similar to images captured from real ones. ...
... For instance, the fingerprint scanner in Apple's iPhone 5s was fooled in 2013 by German Hackers group Chaos Computer Club using an counterfeited fingerprint , while a team from University of Hanoi (Vietnam) has demonstrated that how to trick Lenovo, Asus and Toshiba laptops' Face Recognition using genuine user's photograph. Typical countermeasure to spoofing attacks is liveness detection, which aims at detecting physiological signs such as eye blinking and so forth to identify live and artificial biometric traits [2, 3, 4, 5] . Though, several countermeasures against spoofing have been so far proposed, yet none of them have shown to reach a very low error rates. ...
Biometric authentication is now being used ubiquitously as an alternative to passwords on mobile devices. However, current biometric systems are vulnerable to simple spoofing attacks. Several liveness detection methods have been proposed to determine whether there is a live person or an artificial replica in front of the biometric sensor. Yet, the problem is unsolved due to hardship in finding discriminative and computationally inexpensive features for spoofing attacks. Moreover, previous liveness detection approaches are not explicitly aimed for mobile biometric, thus principally unsuited for portable devices. Therefore, we build a software-based multi-biometric prototype that detects face, iris and fingerprint spoofing attacks on mobile devices. We present MoBio-LivDet (Mobile Biometric Liveness Detection), a novel approach that analyzes local features and global structures of the biometric images using a set of low-level feature descriptors and decision level fusion. The system allows user to balance the security level (robustness against spoofing) and convenience that they want. The proposed method is highly fast, simple, efficient, robust and does not require user-cooperation, thus making it extremely apt for mobile devices. Experimental analysis on publicly available face, iris and fingerprint data sets with real spoofing attacks show promising results.
... Spoof attack as stolen, copied or synthetically replicated biometric trait to the sensor to damage the biometric system security in order to gain unauthorized access. An attack on a biometric system can take place for three main reasons [39]: A person may wish to disguise his own identity. For instance, an individual/terrorist attempting to enter a country without legal permission may try to modify his biometric trait or conceal it by placing an artificial biometric trait (e.g. a synthetic fingerprint, mask, or contact lens) over his biometric trait. ...
In identity management system, frequently used biometric recognition system
needs awareness towards issue of protecting biometric template as far as more
reliable solution is apprehensive. In sight of this biometric template
protection algorithm should gratify the basic requirements viz. security,
discriminability and cancelability. As no single template protection method is
capable of satisfying these requirements, a novel scheme for face template
generation and protection is proposed. The novel scheme is proposed to provide
security and accuracy in new user enrolment and authentication process. This
novel scheme takes advantage of both the hybrid approach and the binary
discriminant analysis algorithm. This algorithm is designed on the basis of
random projection, binary discriminant analysis and fuzzy commitment scheme.
Publicly available benchmark face databases (FERET, FRGC, CMU-PIE) and other
datasets are used for evaluation. The proposed novel scheme enhances the
discriminability and recognition accuracy in terms of matching score of the
face images for each stage and provides high security against potential attacks
namely brute force and smart attacks. In this paper, we discuss results viz.
averages matching score, computation time and security for hybrid approach and
novel approach.
This paper introduces DIALERAUTH - a mechanism which leverages the way a smartphone user taps/enters any text-independent 10-digit number (replicating the dialing process) and the hand's micro-movements she makes while doing so. DIALERAUTH authenticates the user on the basis of timing differences in the entered 10-digit strokes. DIALERAUTH provides enhanced security by leveraging the transparent and unobservable layer based on another modality - user's hand micro-movements. Furthermore, DIALERAUTH increases the usability and acceptability by utilizing the users' familiarity with the dialing process and the flexibility of choosing any combination of 10-digit number. We implemented DIALERAUTH for both data collection and proof-of-concept real-time analysis. We collected, in total 10500 legitimate samples involving 97 users, through an extensive unsupervised field experiment, to evaluate the effectiveness of DIALERAUTH. Analysis using one-class Multilayer Perceptron (MLP) shows a True Acceptance Rate (TAR) of 85.77% in identifying the genuine users. Security analysis involving 240 adversarial attempts proved DIALERAUTH as significantly resilient against random and mimic attacks. A usability study based on System Usability Scale (SUS) reflects a positive feedback on user acceptance (SUS score = 73.29).
