Context Switching Algorithm for Selective Multibiometric Fusion
15 Dec 2009-pp 452-457
TL;DR: A multimodal biometric fusion algorithm that supports biometric image quality and case-based context switching approach for selecting appropriate constituent unimodal traits and fusion algorithms is presented.
Abstract: This paper presents a multimodal biometric fusion algorithm that supports biometric image quality and case-based context switching approach for selecting appropriate constituent unimodal traits and fusion algorithms. Depending on the quality of input samples, the proposed algorithm intelligently selects appropriate fusion algorithm for optimal performance. Experiments and correlation analysis on a multimodal database of 320 subjects show that the context switching algorithm improves the verification performance both in terms of accuracy and time.
Citations
More filters
TL;DR: This paper presents an adaptive context switching algorithm coupled with online learning to address the scalability and accommodate the variations in data distribution of biometrics.
27Â citations
Cites methods from "Context Switching Algorithm for Sel..."
...[24] proposed a parallel algorithm to select an appropriate constituent unimodal matcher or the fusion algorithm....
[...]
TL;DR: Experimental results on different multi-modal databases involving face and fingerprint show that the proposed quality-based classifier selection framework yields good performance even when the quality of the biometric sample is sub-optimal.
Abstract: Multibiometric systems fuse the evidence (e.g., match scores) pertaining to multiple biometric modalities or classifiers. Most score-level fusion schemes discussed in the literature require the processing (i.e., feature extraction and matching) of every modality prior to invoking the fusion scheme. This paper presents a framework for dynamic classifier selection and fusion based on the quality of the gallery and probe images associated with each modality with multiple classifiers. The quality assessment algorithm for each biometric modality computes a quality vector for the gallery and probe images that is used for classifier selection. These vectors are used to train Support Vector Machines (SVMs) for decision making. In the proposed framework, the biometric modalities are arranged sequentially such that the stronger biometric modality has higher priority for being processed. Since fusion is required only when all unimodal classifiers are rejected by the SVM classifiers, the average computational time of the proposed framework is significantly reduced. Experimental results on different multi-modal databases involving face and fingerprint show that the proposed quality-based classifier selection framework yields good performance even when the quality of the biometric sample is sub-optimal.
20Â citations
Cites background from "Context Switching Algorithm for Sel..."
...Vatsa et al. [15] proposed a casebased context switching framework for incorporating biometric image quality....
[...]
04 Mar 2021
TL;DR: A framework for multibiometric systems is developed, which combines a deep learning technique with a serial fusion method and improves accuracy by leveraging deep learning technology in feature extraction and score generation.
Abstract: We develop a framework for multibiometric systems, which combines a deep learning technique with the serial fusion method. Deep learning techniques have been used in unimodal and parallel fusion-based multimodal biometric systems in the past few years. While deep learning techniques have been successful in improving the authentication accuracy, a biometric system is still challenged by two issues: 1) a unimodal system suffers from environmental interference, spoofing attacks, and nonuniversality, and 2) a parallel fusion-based multimodal system suffers from user inconvenience as it requires the user to provide multiple biometrics, which in turn takes longer verification times. A serial fusion method can improve user convenience in a multibiometric system by requiring a user to submit only a subset of the available biometrics. To our knowledge, the effectiveness of using a deep learning technique with a serial fusion method in multibiometric systems is still underexplored. In this article, we close this research gap. We develop a three-stage multibiometric system using a user's fingerprint, palm, and face and test three serial fusion methods with a Siamese neural network. Our experiments achieve an AUC of 0.9996, where the genuine users require only 1.56 biometrics (instead of all 3) on an average. Impact statement— We work on enhancing the user convenience and reducing the verification error in a multibiometric system. An improved multibiometric system can help law enforcement, homeland security, defense, and our daily lives by providing better access control. With the advent of deep learning technologies, the accuracy of multibiometric systems have been improved significantly; however, its applicability is still in question because of long verification times required by parallel fusion in a multibiometric system. Our proposed multibiometric framework alleviates this user inconvenience issue by utilizing a serial fusion strategy in decision making and improves accuracy by leveraging deep learning technology in feature extraction and score generation.
14Â citations
Cites background from "Context Switching Algorithm for Sel..."
...Another set of studies developed a context switching algorithm for multimodal biometric fusion and dynamically selected a fusion algorithm depending on the quality of the input images of the biometric modalities [21], [45]–[47]....
[...]
TL;DR: Findings indicate that the proposed approach provides higher accuracy than the unimodal MOC and MOH by 11.29% and 5.12%, respectively, and can authenticate 83.85% of users without auxiliary trait at the expense of only 1.21% lower accuracy compared to parallel fusion.
Abstract: Smart cards are widely used to deploy secure and cost effective identity management systems. Integration of biometrics into the smart card leads to a strong two-factor authentication system through the match on card (MOC) process. Since MOC uses fixed authentication strategies during the life cycle of smart card, this leads to a low performance and high failure to acquire error in uncontrolled noisy environments. To solve this problem, this paper proposes a sequential quality based framework for biometric authentication. In the proposed framework a set of classifiers have been used to manage the workflow of the framework based on the quality of samples. Accordingly, subjects can be dynamically authenticated using MOC and MOH. A multimodal chimera database is used to evaluate this framework. Our findings indicate that the proposed approach provides higher accuracy than the unimodal MOC and MOH by 11.29% and 5.12%, respectively. Furthermore, the proposed framework can authenticate 83.85% of users without auxiliary trait at the expense of only 1.21% lower accuracy compared to parallel fusion, which require acquisition of all traits for entire users. Analysis of the results demonstrates that the proposed approach provides a compromise between accuracy, user convenience, security and system complexity.
14Â citations
10 May 2016
TL;DR: A feature-fusion framework is geared toward improving human identification accuracy for both single and multiple biometrics, which was applied to the face and fingerprint to achieve a 91.11% recognition accuracy when using only a single training sample.
Abstract: The lack of multi-biometric fusion guidelines at the feature-level are addressed in this work. A feature-fusion framework is geared toward improving human identification accuracy for both single and multiple biometrics. The foundation of the framework is the improvement over a state-of-the-art uni-modal biometric verification system, which is extended into a multi-modal identification system. A novel multi-biometric system is thus designed based on the framework, which serves as fusion guidelines for multi-biometric applications that fuse at the feature-level. This framework was applied to the face and fingerprint to achieve a 91.11% recognition accuracy when using only a single training sample. Furthermore, an accuracy of 99.69% was achieved when using five training samples.
9Â citations
Cites background from "Context Switching Algorithm for Sel..."
...Multi-modal biometrics can also be used to solve non-universality and bad input data in well-planned applications by intelligently selecting an appropriate modality [2]....
[...]
References
More filters
Book•
01 Jan 2006
TL;DR: Details multi-modal biometrics and its exceptional utility for increasingly reliable human recognition systems and the substantial advantages of multimodal systems over conventional identification methods.
Abstract: Details multimodal biometrics and its exceptional utility for increasingly reliable human recognition systems. Reveals the substantial advantages of multimodal systems over conventional identification methods.
1,068Â citations
03 Sep 2000
TL;DR: It is shown that dependent classifiers could offer a dramatic improvement over the individual accuracy, however, the relationship between dependency and accuracy of the pool is ambivalent.
Abstract: Independence between individual classifiers is typically viewed as an asset in classifier fusion. We study the limits on the majority vote accuracy when combining dependent classifiers. Q statistics are used to measure the dependence between classifiers. We show that dependent classifiers could offer a dramatic improvement over the individual accuracy. However, the relationship between dependency and accuracy of the pool is ambivalent. A synthetic experiment demonstrates the intuitive result that, in general, negative dependence is preferable.
232Â citations
17 Apr 2006
TL;DR: It is concluded that defocus blur, motion blur, and off-angle are the factors that affect recognition performance the most and a fully automated iris image quality evaluation block is designed that operates in two steps.
Abstract: Iris recognition, the ability to recognize and distinguish individuals by their iris pattern, is the most reliable biometric in terms of recognition and identification performance. However, performance of these systems is affected by poor quality imaging. In this work, we extend previous research efforts on iris quality assessment by analyzing the effect of seven quality factors: defocus blur, motion blur, off-angle, occlusion, specular reflection,
lighting, and pixel-counts on the performance of traditional iris recognition system. We have concluded that defocus blur, motion blur, and off-angle are the factors that affect recognition performance the most. We further designed a fully automated iris image quality evaluation block that operates in two steps. First each factor is estimated individually, then the second step involves fusing the estimated factors by using Dempster-Shafer theory approach to evidential reasoning. The designed block is tested on two datasets, CASIA 1.0 and a dataset collected at WVU. Considerable improvement in recognition performance is demonstrated when removing poor quality images evaluated by our quality metric. The upper bound on processing complexity required to evaluate quality of a single image is O(n2 log n), that of a 2D-Fast Fourier Transform.
208Â citations
"Context Switching Algorithm for Sel..." refers methods in this paper
...Specifically, for computing face image quality score we use quality assessment algorithm describe in [2], RDWT based algorithm [3] is used for computing fingerprint quality score and Dempster Shafer theory based algorithm [4] for iris image quality assessment....
[...]
TL;DR: The proposed PPSVM is a natural and an analytical extension of regular SVMs based on the statistical learning theory and is closer to the Bayes optimal without knowing the distributions.
Abstract: This paper proposes a complete framework of posterior probability support vector machines (PPSVMs) for weighted training samples using modified concepts of risks, linear separability, margin, and optimal hyperplane. Within this framework, a new optimization problem for unbalanced classification problems is formulated and a new concept of support vectors established. Furthermore, a soft PPSVM with an interpretable parameter /spl nu/ is obtained which is similar to the /spl nu/-SVM developed by Scho/spl uml/lkopf et al., and an empirical method for determining the posterior probability is proposed as a new approach to determine /spl nu/. The main advantage of an PPSVM classifier lies in that fact that it is closer to the Bayes optimal without knowing the distributions. To validate the proposed method, two synthetic classification examples are used to illustrate the logical correctness of PPSVMs and their relationship to regular SVMs and Bayesian methods. Several other classification experiments are conducted to demonstrate that the performance of PPSVMs is better than regular SVMs in some cases. Compared with fuzzy support vector machines (FSVMs), the proposed PPSVM is a natural and an analytical extension of regular SVMs based on the statistical learning theory.
114Â citations
"Context Switching Algorithm for Sel..." refers methods in this paper
...Finally, to train SVMs, we use radial basis kernel with kernel parameter of 4 (it our experiments we observe that it yields the best accuracy) and to compute soft labels, we use standard density estimation approach [8]....
[...]
TL;DR: This paper presents a face recognition algorithm that addresses two major challenges: when an individual intentionally alters the appearance and features using disguises, and when limited gallery images are available for recognition.
112Â citations
"Context Switching Algorithm for Sel..." refers methods in this paper
...Further, neural network architecture based Gabor transformation [5] is used to extract facial features, state-of-the-art commercial fingerprint and iris feature extraction and matching tools are used for fingerprint and iris recognition [6]....
[...]