基于特征融合的多模态身份识别方法研究

研究了多模态身份识别问题,结合人脸和掌纹两种不同生理特征,提出了基于特征融合的多模态身份识别方法。对人脸和掌纹图像分别进行Gabor小波、二维主元变换(2DPCA)提取图像特征,根据新的权重算法,结合两种模态的特征,利用最邻近分类器进行分类识别。在AMP、ORL人脸库和Poly-U掌纹图像库中的实验结果表明,两种模态的融合能更多地给出决策分析所需的特征信息相比传统的单一模态的人脸或掌纹识别具有较高的识别率,更具安全性和准确性。     

Performance evaluation of score level fusion in multimodal biometric systems

In a multimodal biometric system, the effective fusion method is necessary for combining information from various single modality systems. In this paper the performance of sum rule-based score level fusion and support vector machines (SVM)-based score level fusion are examined. Three biometric characteristics are considered in this study: fingerprint, face, and finger vein. We also proposed a new robust normalization scheme (Reduction of High-scores Effect normalization) which is derived from min-max normalization scheme. Experiments on four different multimodal databases suggest that integrating the proposed scheme in sum rule-based fusion and SVM-based fusion leads to consistently high accuracy. The performance of simple sum rule-based fusion preceded by our normalization scheme is comparable to another approach, likelihood ratio-based fusion [8] (Nandakumar et al., 2008), which is based on the estimation of matching scores densities. Comparison between experimental results on sum rule-based fusion and SVM-based fusion reveals that the latter could attain better performance than the former, provided that the kernel and its parameters have been carefully selected.     

A user-specific and selective multimodal biometric fusion strategy by ranking subjects

The recognition performance of a biometric system varies significantly from one enrolled user to another. As a result, there is a need to tailor the system to each user. This study investigates a relatively new fusion strategy that is both user-specific and selective. By user-specific, we understand that each user in a biometric system has a different set of fusion parameters that have been tuned specifically to a given enrolled user. By selective, we mean that only a subset of modalities may be chosen for fusion. The rationale for this is that if one biometric modality is sufficiently good to recognize a user, fusion by multimodal biometrics would not be necessary, we advance the state of the art in user-specific and selective fusion in the following ways: (1) provide thorough analyses of (a) the effect of pre-processing the biometric output (prior to applying a user-specific score normalization procedure) in order to improve its central tendency and (b) the generalisation ability of user-specific parameters; (2) propose a criterion to rank the users based solely on a training score dataset in such a way that the obtained rank order will maximally correlate with the rank order that is obtained if it were to be computed on the test set; and, (3) experimentally demonstrate the performance gain of a user-specific and -selective fusion strategy across fusion data sets at different values of "pruning rate" that control the percentage of subjects for whom fusion is not required. Fifteen sets of multimodal fusion experiments carried out on the XM2VTS score-level benchmark database show that even though our proposed user-specific and -selective fusion strategy, its performance compares favorably with the conventional fusion system that considers all information.     

Designing efficient fusion schemes for multimodal biometric systems using face and palmprint

In this paper, we address the problem of designing efficient fusion schemes of complementary biometric modalities such as face and palmprint, which are effectively coded using Log-Gabor transformations, resulting in high dimensional feature spaces. We propose different fusion schemes at match score level and feature level, which we compare on a database of 250 virtual people built from the face FRGC and the palmprint PolyU databases. Moreover, in order to reduce the complexity of the fusion scheme, we implement a particle swarm optimization (PSO) procedure which allows the number of features (identifying a dominant subspace of the large dimension feature space) to be significantly reduced while keeping the same level of performance. Results in both closed identification and verification rates show a significant improvement of 6% in performance when performing feature fusion in Log-Gabor space over the more common optimized match score level fusion method.     

Generation of new points for training set and feature-level fusion in multimodal biometric identification

Multimodal biometrics has gained interest in the recent past due to its improved recognition rate over unibiometric and unimodal systems. Fusion at feature level is considered here for the purpose of recognition. The biometrics considered for fusion are face and iris. Here, new face images along with iris images are generated, and they are included in the training set. Feature-level fusion is incorporated. The recognition rates of the classification algorithm thus obtained are statistically found to be significantly better than the existing feature-level fusion and classification techniques.     

A multimodal biometric test bed for quality-dependent, cost-sensitive and client-specific score-level fusion algorithms

This paper presents a test bed, called the Biosecure DS2 score-and-quality database, for evaluating, comparing and benchmarking score-level fusion algorithms for multimodal biometric authentication. It is designed to benchmark quality-dependent, client-specific, cost-sensitive fusion algorithms. A quality-dependent fusion algorithm is one which attempts to devise a fusion strategy that is dependent on the biometric sample quality. A client-specific fusion algorithm, on the other hand, exploits the specific score characteristics of each enrolled user in order to customize the fusion strategy. Finally, a cost-sensitive fusion algorithm attempts to select a subset of biometric modalities/systems (at a specified cost) in order to obtain the maximal generalization performance. To the best of our knowledge, the BioSecure DS2 data set is the first one designed to benchmark the above three aspects of fusion algorithms. This paper contains some baseline experimental results for evaluating the above three types of fusion scenarios.     

Score normalization in multimodal biometric systems

Multimodal biometric systems consolidate the evidence presented by multiple biometric sources and typically provide better recognition performance compared to systems based on a single biometric modality. Although information fusion in a multimodal system can be performed at various levels, integration at the matching score level is the most common approach due to the ease in accessing and combining the scores generated by different matchers. Since the matching scores output by the various modalities are heterogeneous, score normalization is needed to transform these scores into a common domain, prior to combining them. In this paper, we have studied the performance of different normalization techniques and fusion rules in the context of a multimodal biometric system based on the face, fingerprint and hand-geometry traits of a user. Experiments conducted on a database of 100 users indicate that the application of min–max, z-score, and tanh normalization schemes followed by a simple sum of scores fusion method results in better recognition performance compared to other methods. However, experiments also reveal that the min–max and z-score normalization techniques are sensitive to outliers in the data, highlighting the need for a robust and efficient normalization procedure like the tanh normalization. It was also observed that multimodal systems utilizing user-specific weights perform better compared to systems that assign the same set of weights to the multiple biometric traits of all users.     

Adversarial attacks against mouse- and keyboard-based biometric authentication: black-box versus domain-specific techniques

International Journal of Information Security - Adversarial attacks have recently gained popularity due to their simplicity, impact, and applicability to a wide range of machine learning scenarios....     

Bayesian adaptation for user-dependent multimodal biometric authentication

A novel score-level fusion strategy based on Bayesian adaptation for user-dependent multimodal biometric authentication is presented. In the proposed method, the fusion function is adapted for each user based on prior information extracted from a pool of users. Experimental results are reported using on-line signature and fingerprint verification subsystems on the MCYT real bimodal database. The proposed scheme outperforms both user-independent and user-dependent standard approaches. As compared to non-adapted user-dependent fusion, relative improvements of 80% and 55% are obtained for small and large training set sizes, respectively.     

BiolD: a multimodal biometric identification system

Biometric identification systems, which use physical features to check a person's identity, ensure much greater security than password and number systems. Biometric features such as the face or a fingerprint can be stored on a microchip in a credit card, for example. A single feature, however, sometimes fails to be exact enough for identification. Another disadvantage of using only one feature is that the chosen feature is not always readable. Dialog Communication Systems (DCS AG) developed BioID, a multimodal identification system that uses three different features-face, voice, and lip movement-to identify people. With its three modalities, BioID achieves much greater accuracy than single-feature systems. Even if one modality is somehow disturbed-for example, if a noisy environment drowns out the voice-the ether two modalities still lead to an accurate identification. This article goes into detail about the system functions, explaining the data acquisition and preprocessing techniques for voice, facial, and lip imagery data. The authors also explain the classification principles used for optical features and the sensor fusion options (the combinations of the three results-face, voice, lip movement-to obtain varying levels of security)     

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.     

A comparison of imputation methods for handling missing scores in biometric fusion

Multibiometric systems, which consolidate or fuse multiple sources of biometric information, typically provide better recognition performance than unimodal systems. While fusion can be accomplished at various levels in a multibiometric system, score-level fusion is commonly used as it offers a good trade-off between data availability and ease of fusion. Most score-level fusion rules assume that the scores pertaining to all the matchers are available prior to fusion. Thus, they are not well equipped to deal with the problem of missing match scores. While there are several techniques for handling missing data in general, the imputation scheme, which replaces missing values with predicted values, is preferred since this scheme can be followed by a standard fusion scheme designed for complete data. In this work, the performance of the following imputation methods are compared in the context of multibiometric fusion: K-nearest neighbor (KNN) schemes, likelihood-based schemes, Bayesian-based schemes and multiple imputation (MI) schemes. Experiments on the MSU database assess the robustness of the schemes in handling missing scores at different missing rates. It is observed that the Gaussian mixture model (GMM)-based KNN imputation scheme results in the best recognition accuracy.     

MMU GASPFA: A COTS multimodal biometric database

This paper describes the baseline corpus of a new multimodal biometric database, the MMU GASPFA (Gait–Speech–Face) database. The corpus in GASPFA is acquired using commercial off the shelf (COTS) equipment including digital video cameras, digital voice recorder, digital camera, Kinect camera and accelerometer equipped smart phones. The corpus consists of frontal face images from the digital camera, speech utterances recorded using the digital voice recorder, gait videos with their associated data recorded using both the digital video cameras and Kinect camera simultaneously as well as accelerometer readings from the smart phones. A total of 82 participants had their biometric data recorded. MMU GASPFA is able to support both multimodal biometric authentication as well as gait action recognition. This paper describes the acquisition setup and protocols used in MMU GASPFA, as well as the content of the corpus. Baseline results from a subset of the participants are presented for validation purposes.     

Biosec baseline corpus: A multimodal biometric database

The baseline corpus of a new multimodal database, acquired in the framework of the FP6 EU BioSec Integrated Project, is presented. The corpus consists of fingerprint images acquired with three different sensors, frontal face images from a webcam, iris images from an iris sensor, and voice utterances acquired both with a close-talk headset and a distant webcam microphone. The BioSec baseline corpus includes real multimodal data from 200 individuals in two acquisition sessions. In this contribution, the acquisition setup and protocol are outlined, and the contents of the corpus—including data and population statistics—are described. The database will be publicly available for research purposes by mid-2006.     

多模态生物特征提取及相关性评价综述

生物特征识别是身份认证的重要手段,特征提取技术在其中扮演了关键角色,直接影响识别的结果。随着特征提取技术日趋成熟,学者们逐渐将目光投向了生物特征间的相关性问题。本文以单模态和多模态生物识别中的特征提取方法为研究对象,回顾了人脸与指纹的特征提取方法,分析了基于经验知识的特征分类提取方法以及基于深度学习的计算机逻辑采样提取方法,并从图像处理的角度对单模态与多模态方法进行对比。以当前多模态生物特征提取方法和DNA表达过程为引,提出了不同模态的生物特征之间存在相关性的猜想,以及对这一猜想进行建模的思路。在多模态生物特征提取的基础上,对今后可能有进展的各生物特征之间的相关性建模进行了展望。     

Discriminative multimodal biometric authentication based on quality measures

A novel score-level fusion strategy based on quality measures for multimodal biometric authentication is presented. In the proposed method, the fusion function is adapted every time an authentication claim is performed based on the estimated quality of the sensed biometric signals at this time. Experimental results combining written signatures and quality-labelled fingerprints are reported. The proposed scheme is shown to outperform significantly the fusion approach without considering quality signals. In particular, a relative improvement of approximately 20% is obtained on the publicly available MCYT bimodal database.     

DCCA and DMCCA framework for multimodal biometric system

Multimedia Tools and Applications - Deep networks have been successfully applied to unsupervised feature learning for single modalities such as Text, Image, or Audio. Here, we present the Deep...     

Adapted user-dependent multimodal biometric authentication exploiting general information

A novel adapted strategy for combining general and user-dependent knowledge at the decision level in multimodal biometric authentication is presented. User-independent, user-dependent, and adapted fusion and decision schemes are compared by using a bimodal system based on fingerprint and written signature. The adapted approach is shown to outperform the other strategies considered in this paper. Exploiting available information for training the fusion function is also shown to be better than using existing information for post-fusion trained decisions.