Biometric match score fusion using RVM: A case study in multi-unit iris recognition
16 Jun 2012-pp 65-70
TL;DR: Experimental results on the CASIA-Iris-V4 Thousand database show that RVM provides better accuracy compared to single unit iris recognition and existing fusion algorithms.
Abstract: This paper presents a novel fusion approach to combine scores from different biometric classifiers using Relevance Vector Machine. RVM uses a combination of kernel functions on training data for classification and compared to SVM, it requires significantly reduced number of relevance vectors. The proposed RVM based fusion algorithm is evaluated using a case study on multi-unit iris recognition. Experimental results on the CASIA-Iris-V4 Thousand database show that RVM provides better accuracy compared to single unit iris recognition and existing fusion algorithms. With respect to SVM fusion, it is observed that, the accuracy of RVM and SVM are comparable, however, the time for RVM fusion is significantly reduced.
Citations
More filters
01 Dec 2014
TL;DR: A score level fusion scheme for a multimodal biometric system that uses statistical properties of biometric score distribution and an efficient threshold alignment and range compression scheme for score normalization.
Abstract: This paper proposes a score level fusion scheme for a multimodal biometric system. Accuracy and reliability of a system are improved by utilizing more than one samples. Every matching of a biometric sample with its corresponding biometric sample in the database produces a matching score. There multiple scores from different biometric samples are fused for further utilization. It proposes an efficient threshold alignment and range compression scheme for score normalization. It uses statistical properties of biometric score distribution. The proposed scheme has been tested over a multimodal database which is constructed by using three publicly available database viz. FVC2006-DB2-A of fingerprint, CASIA-V4-Lamp of iris and PolyU of palmprint. Experimental results have shown the significant performance boost.
7 citations
Cites methods from "Biometric match score fusion using ..."
...Relevance Vector Machine is used in [12] and a ranking based user specific fusion strategy is proposed in [15]....
[...]
TL;DR: Local Hybrid Binary Gradient Contour and Hierarchical Local Binary Pattern are proposed as the texture descriptors for finger vein recognition to increase the discriminant capability of the finger vein texture.
Abstract: Finger vein recognition is a type of biometric technology that uses the vein pattern inside the human finger as a personal identifier. In this paper, Local Hybrid Binary Gradient Contour (LHBGC) and Hierarchical Local Binary Pattern (HLBP) are proposed as the texture descriptors for finger vein recognition to increase the discriminant capability of the finger vein texture. LHBGC extracts both sign and magnitude components of the finger vein image for recognition, while HLBP utilizes the LBP uniform texture pattern of the vein image without any training required. Furthermore, a multi-instance biometrics that fuses multiple evidences from an individual has also been proposed to address the problem of noisy data. Multi-instance biometrics is the most inexpensive way to obtain multiple biometric evidences from a biometric trait without multiple sensors and additional feature extraction algorithms. Experiments on several benchmark databases validate the efficiency of the proposed multi-instance approach. An equal error rate as low as 0.00002% is achieved using the combination of three fingers at score level fusion.
4 citations
01 Jan 2015
TL;DR: This chapter presents a score level fusion scheme for multimodal biometric system where multiple scores corresponding to the matchings of different biometric samples are fused for taking decision on similarity.
Abstract: This chapter presents a score level fusion scheme for multimodal biometric system. There multiple scores corresponding to the matchings of different biometric samples are fused for taking decision on similarity. Proposed score normalization is an threshold alignment and range compression scheme. It utilizes statistical properties of the score distribution. The proposed scheme has been tested over a multimodal database which is constructed using three publicly available database. Experimental results have shown the significant performance boost.
3 citations
Dissertation•
01 Jan 2018
TL;DR: A novel performance anchored score normalization technique is presented that aligns certain performance-related score values in the fused biometric sources leading to more accurate multi-biometric decisions when compared to conventional normalization approaches.
Abstract: Biometric recognition is the automated recognition of individuals based on their behavioral or biological characteristics. Beside forensic applications, this technology aims at replacing the outdated and attack prone, physical and knowledge-based, proofs of identity. Choosing one biometric characteristic is a tradeoff between universality, acceptability, and permanence, among other factors. Moreover, the accuracy cap of the chosen characteristic may limit the scalability and usability for some applications. The use of multiple biometric sources within a unified frame, i.e. multi-biometrics, aspires to tackle the limitations of single source biometrics and thus enables a wider implementation of the technology. This work aims at presenting application-driven advances in multi-biometrics by addressing different elements of the multi-biometric system work-flow.
At first, practical oriented pre-fusion issues regarding missing data imputation and score normalization are discussed. This includes presenting a novel performance anchored score normalization technique that aligns certain performance-related score values in the fused biometric sources leading to more accurate multi-biometric decisions when compared to conventional normalization approaches. Missing data imputation within score-level multi-biometric fusion is also addressed by analyzing the behavior of different approaches under different operational scenarios.
Within the multi-biometric fusion process, different information sources can have different degrees of reliability. This is usually influenced in the fusion process by assigning relative weights to the fused sources. This work presents a number of weighting approaches aiming at optimizing the decision made by the multi-biometric system. First, weights that try to capture the overall performance of the biometric source, as well as an indication of its confidence, are proposed and proved to outperform the state-of-the-art weighting approaches. The work also introduces a set of weights derived from the identification performance representation, the cumulative match characteristics. The effect of these weights is analyzed under the verification and identification scenarios.
To further optimize the multi-biometric process, information besides the similarity between two biometric captures can be considered. Previously, the quality measures of biometric captures were successfully integrated, which requires accessing and processing raw captures. In this work, supplementary information that can be reasoned from the comparison scores are in focus. First, the relative relation between different biometric comparisons is discussed and integrated in the fusion process resulting in a large reduction in the error rates. Secondly, the coherence between scores of multi-biometric sources in the same comparison is defined and integrated into the fusion process leading to a reduction in the error rates, especially when processing noisy data.
Large-scale biometric deployments are faced by the huge computational costs of running biometric searches and duplicate enrollment checks. Data indexing can limit the search domain leading to faster searches. Multi-biometrics provides richer information that can enhance the retrieval performance. This work provides an optimizable and configurable multi-biometric data retrieval solution that combines and enhances the robustness of rank-level solutions and the performance of feature-level solutions.
Furthermore, this work presents biometric solutions that complement and utilize multi-biometric fusion. The first solution captures behavioral and physical biometric characteristics to assure a continuous user authentication. Later, the practical use of presentation attack detection is discussed by investigating the more realistic scenario of cross-database evaluation and presenting a state-of-the-art performance comparison. Finally, the use of multi-biometric fusion to create face references from videos is addressed. Face selection, feature-level fusion, and score-level fusion approaches are evaluated under the scenario of face recognition in videos.
3 citations
01 Dec 2014
TL;DR: A single hidden-layer feedforward fusion network is proposed for face identity verification and shows that the proposed model consistently outperforms competing methods.
Abstract: In this paper, a single hidden-layer feedforward fusion network is proposed for face identity verification. Essentially, the feature extraction, matching score calculation and fusion algorithm design steps are integrated and absorbed into a hidden layer of the model. Each hidden node works on the raw face image directly and produces an Euclidean distance based match score within the network. These scores are then incorporated with output weights to produce a fused score at the final stage. Our experimental study conducted using three face databases shows that the proposed model consistently outperforms competing methods.
1 citations
Cites background from "Biometric match score fusion using ..."
...Many studies have shown that match scores fusion with an appropriate tuning improves the accuracy [9], [10], [11]....
[...]
...Also, the fusion can usually enhance the veri cation accuracy [9], [10]....
[...]
References
More filters
Book•
01 Jan 1995TL;DR: Setting of the learning problem consistency of learning processes bounds on the rate of convergence ofLearning processes controlling the generalization ability of learning process constructing learning algorithms what is important in learning theory?
Abstract: Setting of the learning problem consistency of learning processes bounds on the rate of convergence of learning processes controlling the generalization ability of learning processes constructing learning algorithms what is important in learning theory?.
40,147 citations
"Biometric match score fusion using ..." refers methods in this paper
...SVM [15] is a widely used classification technique that avoids over-fitting and leads to good generalization by finding the separating hyperplane that maximizes the margin width....
[...]
TL;DR: A common theoretical framework for combining classifiers which use distinct pattern representations is developed and it is shown that many existing schemes can be considered as special cases of compound classification where all the pattern representations are used jointly to make a decision.
Abstract: We develop a common theoretical framework for combining classifiers which use distinct pattern representations and show that many existing schemes can be considered as special cases of compound classification where all the pattern representations are used jointly to make a decision. An experimental comparison of various classifier combination schemes demonstrates that the combination rule developed under the most restrictive assumptions-the sum rule-outperforms other classifier combinations schemes. A sensitivity analysis of the various schemes to estimation errors is carried out to show that this finding can be justified theoretically.
5,670 citations
"Biometric match score fusion using ..." refers background in this paper
...Kittler et al. developed a theoretical framework for combining the classifiers which includes product rule, sum rule, max rule, min rule, median rule and majority voting [11]....
[...]
TL;DR: It is demonstrated that by exploiting a probabilistic Bayesian learning framework, the 'relevance vector machine' (RVM) can derive accurate prediction models which typically utilise dramatically fewer basis functions than a comparable SVM while offering a number of additional advantages.
Abstract: This paper introduces a general Bayesian framework for obtaining sparse solutions to regression and classification tasks utilising models linear in the parameters Although this framework is fully general, we illustrate our approach with a particular specialisation that we denote the 'relevance vector machine' (RVM), a model of identical functional form to the popular and state-of-the-art 'support vector machine' (SVM) We demonstrate that by exploiting a probabilistic Bayesian learning framework, we can derive accurate prediction models which typically utilise dramatically fewer basis functions than a comparable SVM while offering a number of additional advantages These include the benefits of probabilistic predictions, automatic estimation of 'nuisance' parameters, and the facility to utilise arbitrary basis functions (eg non-'Mercer' kernels) We detail the Bayesian framework and associated learning algorithm for the RVM, and give some illustrative examples of its application along with some comparative benchmarks We offer some explanation for the exceptional degree of sparsity obtained, and discuss and demonstrate some of the advantageous features, and potential extensions, of Bayesian relevance learning
5,116 citations
TL;DR: Algorithms developed by the author for recognizing persons by their iris patterns have now been tested in many field and laboratory trials, producing no false matches in several million comparison tests.
Abstract: Algorithms developed by the author for recognizing persons by their iris patterns have now been tested in many field and laboratory trials, producing no false matches in several million comparison tests. The recognition principle is the failure of a test of statistical independence on iris phase structure encoded by multi-scale quadrature wavelets. The combinatorial complexity of this phase information across different persons spans about 249 degrees of freedom and generates a discrimination entropy of about 3.2 b/mm/sup 2/ over the iris, enabling real-time decisions about personal identity with extremely high confidence. The high confidence levels are important because they allow very large databases to be searched exhaustively (one-to-many "identification mode") without making false matches, despite so many chances. Biometrics that lack this property can only survive one-to-one ("verification") or few comparisons. The paper explains the iris recognition algorithms and presents results of 9.1 million comparisons among eye images from trials in Britain, the USA, Japan, and Korea.
2,829 citations