Search

In this article, we introduce an analytic formulation for compressive binary classification. The formulation seeks to solve the least ℓ p -norm of the parameter vector subject to a classification error constraint. An analytic and stretchable estimation is conjectured where the estimation can be viewed as an extension of the pseudoinverse with left and right constructions. Our variance analysis indicates that the estimation based on the left pseudoinverse is unbiased and the estimation based on the right pseudoinverse is biased. Sparseness can be obtained for the biased estimation under certain mild conditions. The proposed estimation is investigated numerically using both synthetic and real-world data. [ABSTRACT FROM AUTHOR]

A center sliding Bayesian design adopting orthogonal polynomials for binary pattern classification is studied in this paper. Essentially, a Bayesian weight solution is coupled with a center sliding scheme in feature space which provides an easy tuning capability for binary classification. The proposed method is compared with several state-of-the-art binary classifiers in terms of their solution forms, decision thresholds and decision boundaries. Based on the center sliding Bayesian framework, a novel orthogonal polynomial classifier is subsequently developed. The orthogonal polynomial classifier is evaluated using two representative orthogonal polynomials for feature mapping. Our experimental results show promising potential of the orthogonal polynomial classifier since it achieves both desired accuracy and computational efficiency. [ABSTRACT FROM AUTHOR]

Abstract: The structural resemblance among several existing classifiers has motivated us to investigate their underlying relationships. By exploring into the mapping solutions of these classifiers, we found that they can be linked by simple feature data scaling. In other words, the key to these relationships lies upon how the replica of feature data are being scaled. This finding leads us directly to an exploration of novel classifiers beyond existing settings. Based on an extensive empirical evaluation, we show that the proposed formulation facilitates a tuning capability beyond existing settings for classifier generalization. [Copyright &y& Elsevier]

Abstract: In design of a multibiometric system, a major concern is the learning cost in terms of computation complexity and memory usage due to large data size. In this paper, we propose an online learning network to circumvent the computational problem. Although conventional online learning algorithms can be adopted, their optimization of the fitting distance residuals does not meet the actual classification error requirement. A direct optimization to the classification performance is thus desired. Since the proposed classification-based formulation involves a class-specific weight which varies according to the total number of genuine-users and imposters, an online learning formulation becomes non-trivial. Extensive empirical evaluations on publicly available data sets show promising potential of the proposed method in terms of fusion verification accuracy and computational cost. [Copyright &y& Elsevier]

Abstract: In this paper, we propose to extract localized random features directly from partial face image matrix for cancelable identity verification. Essentially, the extracted random features consist of compressed horizontal and vertical facial information obtained from a structured projection of the raw face images. For template security reason, the face appearance information is concealed via averaging several templates over different transformations. The match score outputs of these cancelable templates are then fused through a total error rate minimization. Extensive experiments were carried out to evaluate and benchmark the performance of the proposed method based on the AR, FERET, ORL, Sheffield and BERC databases. Our empirical results show encouraging performances in terms of verification accuracy as well as satisfying four cancelable biometric properties. [Copyright &y& Elsevier]

We propose a service-oriented architecture based on biometric system where training and classification tasks are used by millions of users via internet connection. Such a large-scale biometric system needs to consider template protection, accuracy and efficiency issues. This is a challenging problem since there are tradeoffs among these three issues. In order to simultaneously handle these issues, we extract both global and local features via controlling the sparsity of random bases without training. Subsequently, the extracted features are fused with a sequential classifier. In the proposed system, the random basis features are not stored for security reason. The non-training based on feature extraction followed by a sequential learning contributes to computational efficiency. The overall accuracy is consequently improved via an ensemble of classifiers. We evaluate the performance of the proposed system using equal error rate under a stolen-token scenario. Our experimental results show that the proposed method is robust over severe local deformation with efficient computation for simultaneous transactions. Although we focus on face biometrics in this paper, the proposed method is generic and can be applied to other biometric traits. [ABSTRACT FROM AUTHOR]

Abstract: Due to the rapid growth of social network services such as Facebook and Twitter, incorporation of face recognition in these large-scale web services is attracting much attention in both academia and industry. The major problem in such applications is to deal efficiently with the growing number of samples as well as local appearance variations caused by diverse environments for the millions of users over time. In this paper, we focus on developing an incremental face recognition method for Twitter application. Particularly, a data-independent feature extraction method is proposed via binarization of a Gabor filter. Subsequently, the dimension of our Gabor representation is reduced considering various orientations at different grid positions. Finally, an incremental neural network is applied to learn the reduced Gabor features. We apply our method to a novel application which notifies new photograph uploading to related users without having their ID being identified. Our extensive experiments show that the proposed algorithm significantly outperforms several incremental face recognition methods with a dramatic reduction in computational speed. This shows the suitability of the proposed method for a large-scale web service with millions of users. [Copyright &y& Elsevier]

Abstract: The area under the ROC curve (AUC) provides a good scalar measure of ranking performance without requiring a specific threshold for performance comparison among classifiers. AUC is useful for imprecise environments since it operates independently with respect to class distributions and misclassification costs. A direct optimization of this AUC criterion thus becomes a natural choice for binary classifier design. However, a direct formulation based on the AUC criterion would require a high computational cost due to the drastically increasing input pair features. In this paper, we propose an online learning algorithm to circumvent this computational problem for binary classification. Different from those conventional recursive formulations, the proposed formulation involves a pairwise cost function which pairs up a newly arrived data point with those of opposite class in stored data. Moreover, with incorporation of a sparse learning into the online formulation, the computational effort can be significantly reduced. Our empirical results on three different scales of public databases show promising potential in terms of classification AUC, accuracy, and computational efficiency. [Copyright &y& Elsevier]

Abstract: With an aim of extracting robust facial features under pose variations, this paper presents two directional projections corresponding to extraction of vertical and horizontal local face image features. The matching scores computed from both horizontal and vertical features are subsequently fused at score level via an extreme learning machine that optimizes the total error rate for performance enhancement. In order to benchmark the performance, both the feature extraction and fusion results are compared with that of popular face recognition methods such as principal components analysis and linear discriminant analysis in terms of equal error rate and CPU time. Our empirical experiments using four data sets show encouraging results under considerable horizontal pose variations. [Copyright &y& Elsevier]

Abstract: Due to the increases in processing power and storage capacity of mobile devices over the years, an incorporation of realtime face recognition to mobile devices is no longer unattainable. However, the possibility of the realtime learning of a large number of samples within mobile devices must be established. In this paper, we attempt to establish this possibility by presenting a realtime training algorithm in mobile devices for face recognition related applications. This is differentiated from those traditional algorithms which focused on realtime classification. In order to solve the challenging realtime issue in mobile devices, we extract local face features using some local random bases and then a sequential neural network is trained incrementally with these features. We demonstrate the effectiveness of the proposed algorithm and the feasibility of its application in mobile devices through empirical experiments. Our results show that the proposed algorithm significantly outperforms several popular face recognition methods with a dramatic reduction in computational speed. Moreover, only the proposed method shows the ability to train additional samples incrementally in realtime without memory failure and accuracy degradation using a recent mobile phone model. [ABSTRACT FROM AUTHOR]

Abstract: The receiver operating characteristics (ROC) curve has been extensively used for performance evaluation in multimodal biometrics fusion. However, the processes of fusion classifier design and the final ROC performance evaluation are usually conducted separately. This has been inevitable because the ROC, when taken from the error counting point of view, does not have a well-posed structure linking to the fusion classifier of interest. In this work, we propose to optimize the ROC performance directly according to the fusion classifier design. The area under the ROC curve (AUC) will be used as the optimization objective since it provides a good representation of the ROC performance. Due to the piecewise cumulative structure of the AUC, a smooth approximate formulation is proposed. This enables a direct optimization of the AUC with respect to the classifier parameters. When a fusion classifier has linear parameters, computation of the solution to optimize a quadratic AUC approximation is surprisingly simple and yet effective. Our empirical experiments on biometrics fusion show strong evidences regarding the potential of the proposed method. [Copyright &y& Elsevier]

This letter presents a minimum classification error learning formulation for a single-layer feedforward network (SLFN). By approximating the nonlinear counting step function using a quadratic function, the classification error rate is shown to be deterministically solvable. Essentially the derived solution is related to an existing weighted least-squares method with class-specific weights set according to the size of data set. By considering the class-specific weights as adjustable parameters, the learning formulation extends the classification robustness of the SLFN without sacrificing its intrinsic advantage of being a closed-form algorithm. While the method is applicable to other linear formulations, our empirical results indicate SLFN's effectiveness on classification generalization. [ABSTRACT FROM AUTHOR]

Abstract: This paper proposes a weighted power series model for face verification scores fusion. Essentially, a linear parametric power series model is adopted to directly minimize an approximated total error rate for fusion of multi-modal face verification scores. Unlike the conventional least-squares error minimization approach which involves fitting of a learning model to data density and then perform a threshold process for error counting, this work directly formulates the required target error count rate in terms of design model parameters with a closed-form solution. The solution is found to belong to a specific setting of the weighted least squares. Our experiments on fusing scores from visual and infra-red face images as well as on public data sets show promising results. [Copyright &y& Elsevier]

Abstract: This paper addresses the biometric scores fusion problem from the error rate minimization point of view. Comparing to the conventional approach which treats fusion classifier design and performance evaluation as a two-stage process, this work directly optimizes the target performance with respect to fusion classifier design. Based on a smooth approximation to the total error rate of identity verification, a deterministic solution is proposed to solve the fusion optimization problem. The proposed method is applied to a face and iris verification fusion problem addressing the demand for high security in the modern networked society. Our empirical evaluations show promising potential in terms of decision accuracy and computing efficiency. [Copyright &y& Elsevier]

Abstract: This paper presents a novel quadratic error-counting network for pattern classification. Two computational issues namely, the network learning issue and the classification error-counting issue have been addressed. Essentially, a linear series functional approximation to network structure and a smooth quadratic error-counting cost function were proposed to resolve these two computational issues within a single framework. Our analysis shows that the quadratic error-counting objective can be related to the least-squares-error objective by adjusting the class-specific normalization factors. The binary classification network is subsequently extended to cater for multicategory problems. An extensive empirical evaluation validates the usefulness of proposed method. [Copyright &y& Elsevier]

Abstract: This work addresses the problem of deformable hand shape recognition in biometric systems without any positioning aids. We separate and recognize multiple rigid fingers under Euclidean transformations. An elliptical model is introduced to represent fingers and accelerate the search of optimal alignments of fingers. Unlike other methods, the similarity is measured during alignment search based on finger width measurements defined at nodes by controllable intervals to achieve balanceable recognition accuracy and computational cost. Technically, our method bridges the traditional handcrafted-feature methods and the shape-distance methods. We have tested it using our 108-person-540-sample database with significantly increased positive recognition accuracy. [Copyright &y& Elsevier]

In this paper, we address the multimodal biometric decision fusion problem. By exploring into the user-specific approach for learning and threshold setting, four possible paradigms for learning and decision making are investigated. Since each user requires a decision hyperplane specific to him in order to achieve good verification accuracy, those tedious iterative training methods like the neural network approach would not be suitable. We propose to use a model that requires only a single training step for this application. The four global and local learning and decision paradigms are then explored to observe their decision capability. Besides the proposal of a relevant receiver operating characteristic performance for the local decision, extensive experiments were conducted to observe the verification performance for fusion of two and three biometrics.

A novel method using a reduced multivariate polynomial model has been developed for biometric decision fusion where simplicity and ease of use could be a concern. However, much to our surprise, the reduced model was found to have good classification accuracy for several commonly used data sets from the Web. In this paper, we extend the single output model to a multiple outputs model to handle multiple class problems. The method is particularly suitable for problems with small number of features and large number of examples. Basic component of this polynomial model boils down to construction of new pattern features which are sums of the original features and combination of these new and original features using power and product terms. A linear regularized least-squares predictor is then built using these constructed features. The number of constructed feature terms varies linearly with the order of the polynomial, instead of having a power law in the case of full multivariate polynomials. The method is simple as it amounts to only a few lines of Matlab code. We perform extensive experiments on this reduced model using 42 data sets. Our results compared remarkably well with best reported results of several commonly used algorithms from the literature. Both the classification accuracy and efficiency aspects are reported for this reduced model. [ABSTRACT FROM AUTHOR]

The multivariate polynomial model provides an effective way to describe complex nonlinear input-output relationships since it is tractable for optimization, sensitivity analysis, and prediction of confidence intervals. However, for high-dimensional and high-order problems, multivariate polynomial regression becomes impractical due to its huge number of product terms. This is especially true for the case of a full interaction model. In this paper, we propose a reduced multivariate polynomial model to circumvent the dimensionality problem with some compromise in its approximation capability. In multimodal biometrics and many classifiers fusion applications, as individual classifiers to be combined would have attained a certain level of classification accuracy, this reduced multivariate polynomial model can be used to combine these classifiers in the next level of classification taking their outputs as the inputs to the reduced multivariate polynomial model. The model is first applied to a well-known pattern classification problem to illustrate its classification capability. The reduced multivariate polynomial model is then applied to combine two biometric verification systems with improved receiver operating characteristics performance as compared to an optimal weighing method and a few commonly used classifiers. [ABSTRACT FROM AUTHOR]

This paper addresses the issue of training feedforward neural networks by global optimization. The main contributions include characterization of global optimality of a network error function, and formulation of a global descent algorithm to solve the network training problem. A network with a single hidden-layer and a single-output unit is considered. By means of a monotonic transformation, a sufficient condition for global optimality of a network error function is presented. Based on this, a penaity-based algorithm is derived directing the search towards possible regions containing the global minima. Numerical comparison with benchmark problems from the neural network literature shows superiority of the proposed algorithm over some local methods, in terms of the percentage of trials attaining the desired solutions. The algorithm is also shown to be effective for several pattern recognition problems. [ABSTRACT FROM AUTHOR]

A method based on non-linear regression has been proposed by Toh and Devanathan (1996) to perform closed-loop process identification under naturally occurring load disturbances. The significance of this method includes relaxation of an important assumption on the knowledge of the input excitation source, given a stable controller which is to be improved. While not aiming to identify a precise model as in classical design, this method uses an approximate process model for on-line identification. This serves as a basis for controller refinement, without having to interrupt the process operation for identification test. In this paper, we present a refined formulation and then perform convexity analysis and quasiconvexity analysis on this refined formulation. We show that these results can be utilized to obtain convergence independent of the initial estimates. Applicability of the identification for controller tuning is also demonstrated using a process with an unknown model. [ABSTRACT FROM AUTHOR]

Training neural networks with the Moore–Penrose (MP) inverse has recently gained attention in view of its noniterative training nature. However, a significant drawback of learning based on the MP inverse is that the computational memory consumption grows along with the size of a dataset. In this article, based on the partitioning of the MP inverse, we propose a blockwise recursive MP inverse formulation (BRMP) for network learning with low-memory property while preserving its training effectiveness. The BRMP is an equivalent formulation to its batchwise counterpart since neither approximation nor assumption is made in the derivation process. Our further exploration of this recursive method leads to a switching structure among three different scenarios. This structure also reveals that the well-known recursive least squares method is a special case of our proposed technique. Subsequently, we apply BRMP to the training of radial basis function networks as well as multilayer perceptrons. The experimental validation covers both regression and classification tasks. [ABSTRACT FROM AUTHOR]

In this paper, we propose a recursive learning formulation for online heterogeneous face recognition (HFR). The main task is to compare between images which are acquired from different sensing spectrums for identity recognition. Using an extreme learning machine, the proposed recursive formulation seeks a direct optimization to the classification error goal where the solution converges exactly to the batch mode solution. Due to the nonlinear nature of the classification error objective function, formulation of a recursive solution that converges is an important and nontrivial task. Based on this recursive formulation, an online HFR system is designed. The system is evaluated using two challenging heterogeneous face databases with images captured under visible, near infrared and infrared spectrums. The proposed system shows promising performance which is comparable with that of competing state-of-the-arts. [ABSTRACT FROM AUTHOR]

In this article, we report the use of randomly structured light illumination for chemical imaging of molecular distribution based on Raman microscopy with improved image resolution. Random structured basis images generated from temporal and spectral characteristics of the measured Raman signatures were superposed to perform structured illumination microscopy (SIM) with the blind-SIM algorithm. For experimental validation, Raman signatures corresponding to Rhodamine 6G (R6G) in the waveband of 730–760 nm and Raman shift in the range of 1096–1634 cm−1 were extracted and reconstructed to build images of R6G. The results confirm improved image resolution using the concept and hints at super-resolution by almost twice better than the diffraction-limit. [ABSTRACT FROM AUTHOR]

Face photo-sketch recognition plays an important role in law enforcement, particularly in narrowing down the search for potential suspects based on limited sketch information. However, the issues of large modality gap and having a relatively small number of sketch samples for training remained a challenging task. In this paper, we propose a novel feature descriptor network for automated face photo-sketch recognition that is suitable for modality discrepancy and small dataset learning. By stacking a multi-directional image difference operation over a pooling projection in a multilayer fashion, our proposal forms an interpretable learning system that does not show obvious overfitting on limited training data. Extensive evaluation using three public face photo-sketch databases shows competing rank-1 recognition accuracy of the proposed method comparing with state-of-the-art methods. In terms of average ranking on the three experimented databases, the proposed method has the top average rank of 2 among 17 algorithms with the runner-up LFDA algorithm having an average rank of 2.83. [ABSTRACT FROM AUTHOR]

Surface plasmon resonance microscopy (SPRM) combines the principles of traditional microscopy with the versatility of surface plasmons to develop label-free imaging methods. This paper describes a proof-of-principles approach based on deep learning that utilized the Y-Net convolutional neural network model to improve the detection and analysis methodology of SPRM. A machine-learning based image analysis technique was used to provide a method for the one-shot analysis of SPRM images to estimate scattering parameters such as the scatterer location. The method was assessed by applying the approach to SPRM images and reconstructing an image from the network output for comparison with the original image. The results showed that deep learning can localize scatterers and predict other variables of scattering objects with high accuracy in a noisy environment. The results also confirmed that with a larger field of view, deep learning can be used to improve traditional SPRM such that it localizes and produces scatterer characteristics in one shot, considerably increasing the detection capabilities of SPRM. [ABSTRACT FROM AUTHOR]

Wi-Fi signals are ubiquitous and provide a convenient, covert, and non-invasive means of recognizing human activity, which is particularly useful for healthcare monitoring. In this study, we investigate a score-level fusion structure for human activity recognition using the Wi-Fi channel state information (CSI) signals. The raw CSI signals undergo an important preprocessing stage before being classified using conventional classifiers at the first level. The output scores of two conventional classifiers are then fused via an analytic network that does not require iterative search for learning. Our experimental results show that the fusion provides good generalization and a shorter learning processing time compared with state-of-the-art networks. [ABSTRACT FROM AUTHOR]

With the capability of fusing varying features from a specific image region, the Region Covariance Matrices (RCM) image descriptor has been evidenced plausible in face recognition. However, a systematic study for RCM, regarding which features to be fused in particular, remains absent. This paper therefore explores several features derived from the orthogonal filter ensembles, i.e., Identity Transform, Discrete Haar Transform, Discrete Cosine Transform, and Karhunen-Loève Transform, for feature encoding in RCM. Aside from that, we also outline a RCM variant, dubbed Region Log-TiedRank Covariance Matrices (RLTCM) in this paper. The RLTCM descriptor, on average, exhibits dramatic performance gain over RCM as well as state-of-the-art descriptors, especially when probe sets far deviated from the face gallery. Furthermore, we discern that the RLTCM descriptor defined based on Identity Transform, i.e., the simplest form of orthogonal filters, and other learning-free orthogonal filters yield impressive performance on par with the learning-based counterparts. [ABSTRACT FROM AUTHOR]

The accurate location of eyes in a facial image is important to many human facial recognition-related applications, and has attracted considerable research interest in computer vision. However, most prevalent methods are based on the frontal pose of the face, where applying them to non-frontal poses can yield erroneous results. In this paper, we propose an eye detection method that can locate the eyes in facial images captured at various head poses. Our proposed method consists of two stages: eye candidate detection and eye candidate verification. In eye candidate detection, eye candidates are obtained by using multi-scale iris shape features and integral image. The size of the iris in face images varies as the head pose changes, and the proposed multi-scale iris shape feature method can detect the eyes in such cases. Since it utilizes the integral image, its computational cost is relatively low. The extracted eye candidates are then verified in the eye candidate verification stage using a support vector machine (SVM) based on the feature-level fusion of a histogram of oriented gradients (HOG) and cell mean intensity features. We tested the performance of the proposed method using the Chinese Academy of Sciences' Pose, Expression, Accessories, and Lighting (CAS-PEAL) database and the Pointing'04 database. The results confirmed the superiority of our method over the conventional Haar-like detector and two hybrid eye detectors under relatively extreme head pose variations. [ABSTRACT FROM AUTHOR]

The iris recognition on a mobile phone is different from the conventional iris recognition implemented on a dedicated device in that the computational power of a mobile phone and the space for placing NIR (near infrared) LED (light emitting diode) illuminators and iris camera are limited. This paper raises these issues in detail based on real implementation of an iris recognition system in a mobile phone and proposes some solutions to these issues. An experimental study was conducted to search for the relevant power and wavelength of NIR LED illuminators with their positioning on a phone for capturing a good quality iris image. Subsequently, in view of the disparity between the user's gazing point and the center of the iris camera which causes degradation of acquired iris images, an experiment was performed to locate the appropriate gazing point for good iris image capture. A fast eye detection algorithm was proposed for implementation under the mobile platform with low computational facility. The experiments were conducted on a currently released mobile phone and the results showed promising potential for adoption of iris recognition as a reliable authentication means. As a result, two 850 nm LEDs were selected for iris illumination at 1.1 cm away from the iris camera for the size of a 7 cm × 13.7 cm phone. In the performance, the recognition accuracy was 0.1% EER (equal error rate) and the eye detection rate with the speed of 17.64 ms on a mobile phone was 99.4%. [ABSTRACT FROM AUTHOR]

A sequential extreme learning machine incorporating a noise compensation scheme via an information measure is developed. In this design, the computationally simple extreme learning machine architecture is maintained while survival error information potential function provides a mechanism for noise compensation. The error compensation is updated online via an error codebook design where an error tolerant and stable solution is obtained. The developed method is tested on chaotic time sequence as well as benchmark data sets. Experimental results show potential applications for the developed method. [ABSTRACT FROM AUTHOR]

This paper presents a visual object tracking system which is tolerant to external imaging factors such as illumination, scale, rotation, occlusion and background changes. Specifically, an integration of an online version of total-error-rate minimization based projection network with an observation model of particle filter is proposed to effectively distinguish between the target object and the background. A re-weighting technique is proposed to stabilize the sampling of particle filter for stochastic propagation. For self-adaptation, an automatic updating scheme and extraction of training samples are proposed to adjust to system changes online. Our qualitative and quantitative experiments on 16 public video sequences show convincing performances in terms of tracking accuracy and computational efficiency over competing state-of-the-art algorithms. [ABSTRACT FROM AUTHOR]

Abstract: In this paper, we propose to combine sclera and periocular features for identity verification. The proposal is particularly useful in applications related to face recognition when the face is partially occluded with only periocular region revealed. Due to its relatively new exposition in the literature of biometrics, particular attention will be paid to sclera feature extraction in this work. For periocular feature extraction, structured random projections were adopted to extract compressed vertical and horizontal components of image features. The binary sclera features are eventually fused with the periocular features at a score level. Extensive experiments have been performed on UBIRIS v1 session1 and session2 databases to assess the verification performance before and after fusion. Around 5% of equal error rate performance was observed to be enhanced by fusing sclera with periocular features comparing with that before fusion. [Copyright &y& Elsevier]

Deep learning (DL) has achieved great success in many applications, but it has been less well analyzed from the theoretical perspective. The unexplainable success of black-box DL models has raised questions among scientists and promoted the emergence of the field of explainable artificial intelligence (XAI). In robotics, it is particularly important to deploy DL algorithms in a predictable and stable manner as robots are active agents that need to interact safely with the physical world. This paper presents an analytic deep learning framework for fully connected neural networks, which can be applied for both regression problems and classification problems. Examples for regression and classification problems include online robot control and robot vision. We present two layer-wise learning algorithms such that the convergence of the learning systems can be analyzed. Firstly, an inverse layer-wise learning algorithm for multilayer networks with convergence analysis for each layer is presented to understand the problems of layer-wise deep learning. Secondly, a forward progressive learning algorithm where the deep networks are built progressively by using single hidden layer networks is developed to achieve better accuracy. It is shown that the progressive learning method can be used for fine-tuning of weights from convergence point of view. The effectiveness of the proposed framework is illustrated based on classical benchmark recognition tasks using the MNIST and CIFAR-10 datasets and the results show a good balance between performance and explainability. The proposed method is subsequently applied for online learning of robot kinematics and experimental results on kinematic control of UR5e robot with unknown model are presented. [ABSTRACT FROM AUTHOR]

Abstract: A biometric discretization scheme converts biometric features into a binary string via segmenting every one-dimensional feature space into multiple labelled intervals, assigning each interval-captured feature element with a short binary string and concatenating the binary output of all feature elements into a bit string. This paper proposes a bit allocation algorithm for biometric discretization to allocate bits dynamically to every feature element based on a Binary Reflected Gray code. Unlike existing bit allocation schemes, our scheme bases upon a combination of bit statistics (reliability measure) and signal to noise ratio (discriminability measure) in performing feature selection and bit allocation procedures. Several empirical comparative studies are conducted extensively on two popular face datasets to justify the efficiency and feasibility of our proposed approach. [Copyright &y& Elsevier]

Abstract: In this paper, we present a novel and effective feature extraction technique for face recognition. The proposed technique incorporates a kernel trick with Graph Embedding and the Fisher’s criterion which we call it as Kernel Discriminant Embedding (KDE). The proposed technique projects the original face samples onto a low dimensional subspace such that the within-class face samples are minimized and the between-class face samples are maximized based on Fisher’s criterion. The implementation of kernel trick and Graph Embedding criterion on the proposed technique reveals the underlying structure of data. Our experimental results on face recognition using ORL, FRGC and FERET databases validate the effectiveness of KDE for face feature extraction. [Copyright &y& Elsevier]

Although the use of biometrics for security access is convenient and easy to be implemented, it also introduced privacy and other security concerns when the original biometric templates are compromised. BioHash was introduced as a form of cancellable or replaceable biometrics through the integration of a set of user-specific random numbers with biometric features to address these concerns. However, the main drawback of the original form of BioHash is its inferior performance when an imposter obtains a legitimate token and uses it to claim as a genuine user (also known as the stolen-token scenario). In this paper, the problem is circumvented by a user-dependent multi-state discretization method. The experimental results on fingerprint database FVC2002 demonstrated a promising performance improvement on the stolen-token scenario when this discretization method was incorporated in the BioHash scheme. Moreover, the discretization method can render a long bit string, which is a useful feature to resist brute-force attacks. Some desired properties such as one-way transformation and diversity are also analyzed. [ABSTRACT FROM AUTHOR]

Abstract: The primary goal of linear discriminant analysis (LDA) in face feature extraction is to find an effective subspace for identity discrimination. The introduction of kernel trick has extended the LDA to nonlinear decision hypersurface. However, there remained inherent limitations for the nonlinear LDA to deal with physical applications under complex environmental factors. These limitations include the use of a common covariance function among each class, and the limited dimensionality inherent to the definition of the between-class scatter. Since these problems are inherently caused by the definition of the Fisher''s criterion itself, they may not be solvable under the conventional LDA framework. This paper proposes to adopt a margin-based between-class scatter and a regularization process to resolve the issue. Essentially, we redesign the between-class scatter matrix based on the SVM margins to facilitate an effective and reliable feature extraction. This is followed by a regularization of the within-class scatter matrix. Extensive empirical experiments are performed to compare the proposed method with several other variants of the LDA method using the FERET, AR, and CMU-PIE databases. [Copyright &y& Elsevier]

Abstract: In this paper, we propose a performance driven methodology for cancelable face templates generation. This is to address the issue of satisfying both the security and performance requirements at the same time. Essentially, the methodology consists of two transformations namely, an efficient feature extraction transformation and an error minimizing template transformation. The first transformation is achieved via a modified sparse random projection which extracts and transforms essential face features into cancelable templates. The second transformation is realized through a direct objective formulation to minimize the system''s total error rate. In order to facilitate convergence of the resulted minimization search, a modified sigmoid is proposed for an error counting step function approximation. Using two publicly available face databases, we empirically show an improved verification performance in terms of the equal error rate while hiding the face identity simultaneously. [Copyright &y& Elsevier]

Highlights • We present a representation of the nose shape. • We propose a nasal similarity measure based on the metric of nose curves in the shape space. • We construct a nasal hierarchical structure for nose statistical analysis. Abstract We propose a novel method for measuring the nasal similarity among 3D faces. Firstly, we construct a representation for the nose shape, which is composed of a set of geodesic curves, each crosses the bridge of the nose. Next, using these geodesic curves, we formulate a similarity measure to compare among noses in the curve shape space. Under the Riemannian framework, the shape space is a quotient space for which the scaling, translation and rotation are removed. Since the nose similarity measure is based on the shape comparison, the proposed method has the following advantages: (1) the similarity measure is robust to facial expressions since the nose is not affected by facial expressions; (2) the geometric features of the nose shape match well with the human perception; (3) the similarity measure is independent of the mesh grid because the chosen nose curves are not sensitive to the triangular mesh model. We construct a nasal hierarchical structure for noses organization which is based on nose similarity measure results. In our experiments, we evaluate the performance of the proposed method and compare it with competing methods on three public face databases namely, FRGC2.0, Texas3D and BosphorusDB. The results show superiority of the proposed method in terms of both the speed and the accuracy when the nasal measurements are processed in the nasal hierarchical structure and the nasal samples with low sampling rate (5%-25% of original point cloud). [ABSTRACT FROM AUTHOR]

In this paper, we propose a single hidden-layer Gabor-based network for heterogeneous face recognition. The proposed input layer contains novel computational units which propagate geometrically localized input image sub-blocks to hidden nodes. The propagated pixels are then convolved with a set of Gabor kernels followed by a randomly weighted summation and a non-linear activation function operation. The output layer adopts a linear weighting scheme which can be deterministically estimated similar to that in extreme learning machine. Our experiments on three experimental scenarios using BERC visual-thermal infrared database and CASIA visual-near infrared database show promising results for the proposed network. [ABSTRACT FROM AUTHOR]

Ever since the availability of real-time three-dimensional (3D) data acquisition sensors such as time-of-flight and Kinect depth sensor, the performance of gesture recognition can be largely enhanced. However, since conventional two-dimensional (2D) image based feature extraction methods such as local binary pattern (LBP) generally use texture information, they cannot be applied to depth or range image which does not contain texture information. In this paper, we propose an adaptive local binary pattern (ALBP) for effective depth images based applications. Contrasting to the conventional LBP which is only rotation invariant, the proposed ALBP is invariant to both rotation and the depth distance in range images. Using ALBP, we can extract object features without using texture or color information. We further apply the proposed ALBP for hand tracking using depth images to show its effectiveness and its usefulness. Our experimental results validate the proposal. [ABSTRACT FROM AUTHOR]

Abstract: We propose a face detection method based on skin color likelihood via a boosting algorithm which emphasizes skin color information while deemphasizing non-skin color information. A stochastic model is adapted to compute the similarity between a color region and the skin color. Both Haar-like features and Local Binary Pattern (LBP) features are utilized to build a cascaded classifier. The boosted classifier is implemented based on skin color emphasis to localize the face region from a color image. Based on our experiments, the proposed method shows good tolerance to face pose variation and complex background with significant improvements over classical boosting-based classifiers in terms of total error rate performance. [Copyright &y& Elsevier]