Multimodal biometrics using geometry preserving projections

Multimodal biometric system utilizes two or more individual modalities, e.g., face, gait, and fingerprint, to improve the recognition accuracy of conventional unimodal methods. However, existing multimodal biometric methods neglect interactions of different modalities during the subspace selection procedure, i.e., the underlying assumption is the independence of different modalities. In this paper, by breaking this assumption, we propose a Geometry Preserving Projections (GPP) approach for subspace selection, which is capable of discriminating different classes and preserving the intra-modal geometry of samples within an identical class. With GPP, we can project all raw biometric data from different identities and modalities onto a unified subspace, on which classification can be performed. Furthermore, the training stage is carried out once and we have a unified transformation matrix to project different modalities. Unlike existing multimodal biometric systems, the new system works well when some modalities are not available. Experimental results demonstrate the effectiveness of the proposed GPP for individual recognition tasks.     

Database-friendly random projections: Johnson-Lindenstrauss with binary coins

A classic result of Johnson and Lindenstrauss asserts that any set of n points in d-dimensional Euclidean space can be embedded into k-dimensional Euclidean space—where k is logarithmic in n and independent of d—so that all pairwise distances are maintained within an arbitrarily small factor. All known constructions of such embeddings involve projecting the n points onto a spherically random k-dimensional hyperplane through the origin. We give two constructions of such embeddings with the property that all elements of the projection matrix belong in {−1,0,+1}. Such constructions are particularly well suited for database environments, as the computation of the embedding reduces to evaluating a single aggregate over k random partitions of the attributes.     

Random multispace quantization as an analytic mechanism for BioHashing of biometric and random identity inputs

Biometric analysis for identity verification is becoming a widespread reality. Such implementations necessitate large-scale capture and storage of biometric data, which raises serious issues in terms of data privacy and (if such data is compromised) identity theft. These problems stem from the essential permanence of biometric data, which (unlike secret passwords or physical tokens) cannot be refreshed or reissued if compromised. Our previously presented biometric-hash framework prescribes the integration of external (password or token-derived) randomness with user-specific biometrics, resulting in bitstring outputs with security characteristics (i.e., noninvertibility) comparable to cryptographic ciphers or hashes. The resultant BioHashes are hence cancellable, i.e., straightforwardly revoked and reissued (via refreshed password or reissued token) if compromised. BioHashing furthermore enhances recognition effectiveness, which is explained in this paper as arising from the random multispace quantization (RMQ) of biometric and external random inputs     

Partial realization of random systems

Consider a vector-valued stationary random process {Y_k}^∞_−∞, from which the estimates, R₀, R₁, …, R_N, of the covariance matrices EY_kY′_k−i, i=0, 1, 1, …, N, can be made.     

结构特征下的可撤销人脸识别

目的 相对于其他生物特征识别技术,人脸识别具有非接触、不易察觉和易于推广等特点,在公共安全和日常生活中得到广泛应用。在移动互联网时代,云端人脸识别可以有效地提高识别精度,但是需要将大量的人脸数据上传到第三方服务器。由于人的面部特征是唯一的,一旦数据库泄露就会面临模板攻击和假冒攻击等安全威胁。为了保证人脸识别系统的安全性并提高其识别率,本文提出一种融合人脸结构特征的可撤销人脸识别算法。方法 首先,对原始人脸图像提取结构特征作为虚部分量,与原始人脸图像联合构建复数矩阵并通过随机二值矩阵进行置乱操作。然后,使用2维主成分分析方法将置乱的复数矩阵映射到新的特征空间。最后,采用基于曼哈顿距离的最近邻分类器计算识别率。结果 在4个不同人脸数据库上的实验结果表明,原始人脸图像和结构特征图像经过随机二值矩阵置乱后,人眼无法察觉出有用的信息且可以重新生成,而且融合方差特征后,在GT （Georgia Tech）、NIR （Near Infrared）、VIS （Visible Light）和YMU （YouTuBe Makeup）人脸数据库上,平均人脸识别率分别提高了4.9%、2.25%、2.25%和1.98%,且平均测试时间均在1.0 ms之内,表明该算法实时性强,能够满足实际应用场景的需求。结论 本文算法可在不影响识别率的情况下保证系统的安全性,满足可撤销性。同时,融合结构特征丰富了人脸信息的表征,提高了人脸识别系统的识别率。     

Sorted random projections for robust rotation-invariant texture classification

This paper presents a simple, novel, yet very powerful approach for robust rotation-invariant texture classification based on random projection. The proposed sorted random projection maintains the strengths of random projection, in being computationally efficient and low-dimensional, with the addition of a straightforward sorting step to introduce rotation invariance. At the feature extraction stage, a small set of random measurements is extracted from sorted pixels or sorted pixel differences in local image patches. The rotation invariant random features are embedded into a bag-of-words model to perform texture classification, allowing us to achieve global rotation invariance. The proposed unconventional and novel random features are very robust, yet by leveraging the sparse nature of texture images, our approach outperforms traditional feature extraction methods which involve careful design and complex steps. We report extensive experiments comparing the proposed method to six state-of-the-art methods, RP, Patch, LBP, WMFS and the methods of Lazebnik et al. and Zhang et al., in texture classification on five databases: CUReT, Brodatz, UIUC, UMD and KTH-TIPS. Our approach leads to significant improvements in classification accuracy, producing consistently good results on each database, including what we believe to be the best reported results for Brodatz, UMD and KTH-TIPS.     

2D compressed learning: support matrix machine with bilinear random projections

Machine Learning - Support matrix machine (SMM) is an efficient matrix classification method that can leverage the structure information within the matrix to improve the classification performance....     

A new method for anomaly detection based on non-convex boundaries with random two-dimensional projections

The implementation of anomaly detection systems represents a key problem that has been focusing the efforts of scientific community. In this context, the use one-class techniques to model a training set of non-anomalous objects can play a significant role. One common approach to face the one-class problem is based on determining the geometric boundaries of the target set. More specifically, the use of convex hull combined with random projections offers good results but presents low performance when it is applied to non-convex sets. Then, this work proposes a new method that face this issue by implementing non-convex boundaries over each projection. The proposal was assessed and compared with the most common one-class techniques, over different sets, obtaining successful results.     

Distributed high dimensional information theoretical image registration via random projections

Information theoretical measures, such as entropy, mutual information, and various divergences, exhibit robust characteristics in image registration applications. However, the estimation of these quantities is computationally intensive in high dimensions. On the other hand, consistent estimation from pairwise distances of the sample points is possible, which suits random projection (RP) based low dimensional embeddings. We adapt the RP technique to this task by means of a simple ensemble method. To the best of our knowledge, this is the first distributed, RP based information theoretical image registration approach. The efficiency of the method is demonstrated through numerical examples.     

基于GMM-WSUM的多生物特征二级融合识别方法

结合基于密度估计和归一化两种融合方法的优点,在匹配分数层级提出了一种基于高斯混合模型(Guassian Mixture Model,GMM)和加权和(Weighted Sums,WSUM)的多生物特征二级融合识别方法。利用GMM对匹配分数建模后,采用N-P准则作为第一级融合策略;第二级融合采用基于加权和的归一化方法,较好地解决了分数归一化融合方法在单模识别算法识别率相差较大时融合识别性能差的问题。在ORL、AR人脸数据库和FVC2004组成的人脸-指纹多模数据库上进行了实验,结果表明,该方法有效地提升了识别性能。     

结合并行融合的序列化多模态生物特征识别系统框架

针对多模态生物特征识别系统并行融合模式中使用方便性和使用效率方面的问题,在现有序列化多模态生物特征识别系统的基础上,提出了一种结合并行融合和序列化融合的多生物特征识别系统框架。框架中首先采用步态、人脸与指纹三种生物特征的不同组合方式以加权相加的得分级融合算法进行的识别过程;其次,利用在线的半监督学习技术提高弱特征的识别性能,从而进一步增强系统的使用方便性和识别可靠性。理论分析和实验结果表明,在此框架下,随使用时间的推移,系统能够通过在线学习提高弱分类器的性能,用户的使用方便性和系统的识别精度都得到了进一步提升。     

A feature extraction method for use with bimodal biometrics

Bimodal biometrics has been found to outperform single biometrics and are usually implemented using the matching score level or decision level fusion, though this fusion will enable less information of bimodal biometric traits to be exploited for personal authentication than fusion at the feature level. This paper proposes matrix-based complex PCA (MCPCA), a feature level fusion method for bimodal biometrics that uses a complex matrix to denote two biometric traits from one subject. The method respectively takes the two images from two biometric traits of a subject as the real part and imaginary part of a complex matrix. MCPCA applies a novel and mathematically tractable algorithm for extracting features directly from complex matrices. We also show that MCPCA has a sound theoretical foundation and the previous matrix-based PCA technique, two-dimensional PCA (2DPCA), is only one special form of the proposed method. On the other hand, the features extracted by the developed method may have a large number of data items (each real number in the obtained features is called one data item). In order to obtain features with a small number of data items, we have devised a two-step feature extraction scheme. Our experiments show that the proposed two-step feature extraction scheme can achieve a higher classification accuracy than the 2DPCA and PCA techniques.     

RNGSSELIB: Program library for random number generation, SSE2 realization

The library RNGSSELIB for random number generators (RNGs) based upon the SSE2 command set is presented. The library contains realization of a number of modern and most reliable generators. Usage of SSE2 command set allows to substantially improve performance of the generators. Three new RNG realizations are also constructed. We present detailed analysis of the speed depending on compiler usage and associated optimization level, as well as results of extensive statistical testing for all generators using available test packages. Fast SSE implementations produce exactly the same output sequence as the original algorithms.

Program summary

Program title: RNGSSELIBCatalogue identifier: AEIT_v1_0Program summary URL:http://cpc.cs.qub.ac.uk/summaries/AEIT_v1_0.htmlProgram obtainable from: CPC Program Library, Queen?s University, Belfast, N. IrelandLicensing provisions: Standard CPC licence, http://cpc.cs.qub.ac.uk/licence/licence.htmlNo. of lines in distributed program, including test data, etc.: 4177No. of bytes in distributed program, including test data, etc.: 21 228Distribution format: tar.gzProgramming language: C.Computer: PC.Operating system: UNIX, Windows.RAM: 1 MbytesClassification: 4.13.Nature of problem: Any calculation requiring uniform pseudorandom number generator, in particular, Monte Carlo calculations.Solution method: The library contains realization of a number of modern and reliable generators: mt19937, mrg32k3a and lfsr113. Also new realizations for the method based on parallel evolution of an ensemble of dynamical systems are constructed: GM19, GM31 and GM61. The library contains both usual realizations and realizations based on SSE command set. Usage of SSE commands allows the performance of all generators to be substantially improved.Restrictions: For SSE realizations of the generators, Intel or AMD CPU supporting SSE2 command set is required. In order to use the realization lfsr113sse, CPU must support SSE4 command set.Running time: Running time is of the order of 20 sec for generating 10⁹ pseudorandom numbers with a PC based on Intel Core i7-940 CPU. Running time is analysed in detail in Section 5 of the paper.     

Rank-one projections with adaptive margins for face recognition.

In supervised dimensionality reduction, tensor representations of images have recently been employed to enhance classification of high dimensional data with small training sets. Previous approaches for handling tensor data have been formulated with tight restrictions on projection directions that, along with convergence issues and the assumption of Gaussian-distributed class data, limit its face-recognition performance. To overcome these problems, we propose a method of rank-one projections with adaptive margins (RPAM) that gives a provably convergent solution for tensor data over a more general class of projections, while accounting for margins between samples of different classes. In contrast to previous margin-based works which determine margin sample pairs within the original high dimensional feature space, RPAM aims instead to maximize the margins defined in the expected lower dimensional feature sub-space by progressive margin refinement after each rank-one projection. In addition to handling tensor data, vector-based variants of RPAM are presented for linear mappings and for nonlinear mappings using kernel tricks. Comprehensive experimental results demonstrate that RPAM brings significant improvement in face recognition over previous subspace learning techniques.     

Combination of hyperbolic functions for multimodal biometrics data fusion.

In this paper, we treat the problem of combining fingerprint and speech biometric decisions as a classifier fusion problem. By exploiting the specialist capabilities of each classifier, a combined classifier may yield results which would not be possible in a single classifier. The Feedforward Neural Network provides a natural choice for such data fusion as it has been shown to be a universal approximator. However, the training process remains much to be a trial-and-error effort since no learning algorithm can guarantee convergence to optimal solution within finite iterations. In this work, we propose a network model to generate different combinations of the hyperbolic functions to achieve some approximation and classification properties. This is to circumvent the iterative training problem as seen in neural networks learning. In many decision data fusion applications, since individual classifiers or estimators to be combined would have attained a certain level of classification or approximation accuracy, this hyperbolic functions network can be used to combine these classifiers taking their decision outputs as the inputs to the network. The proposed hyperbolic functions network model is first applied to a function approximation problem to illustrate its approximation capability. This is followed by some case studies on pattern classification problems. The model is finally applied to combine the fingerprint and speaker verification decisions which show either better or comparable results with respect to several commonly used methods.     

Estimating the viewing parameters of random, noisy projections ofasymmetric objects for tomographic reconstruction

The ability to determine the viewing parameters of objects from their projections has enabled well established tomographic techniques to be employed in the 3D reconstruction of objects from images obtained via modalities where the orientation of the objects cannot be controlled. A method is described for the determination of the viewing parameters of randomly acquired projections of asymmetric objects. It extends upon the common lines algorithm by determining the relative orientation of projections from the location of lines of intersection among the Fourier transforms of the projections in 3D Fourier space. A new technique for finding the lines of intersection in the presence of translational displacement, and for subsequently finding the translational displacement, is presented. The complete algorithm is described and its efficacy is demonstrated using real data. A new technique for dealing with noise is also discussed     

Perceptual video hashing based on temporal wavelet transform and random projections with application to indexing and retrieval of near-identical videos

Multimedia Tools and Applications - A perceptual video hash function extracts a short fixed-length bit string called a perceptual hash on the basis of the visual contents of the video. Such a...