Zhenan Sun

Ordinal palmprint represention for personal identification [represention read representation]

Palmprint-based personal identification, as a new member in the biometrics family, has become an active research topic in recent years. Although great progress has been made, how to represent palmprint for effective classification is still an open problem. In this paper, we present a novel palmprint representation - ordinal measure, which unifies several major existing palmprint algorithms into a general framework. In this framework, a novel palmprint representation method, namely orthogonal line ordinal features, is proposed. The basic idea of this method is to qualitatively compare two elongated, line-like image regions, which are orthogonal in orientation and generate one bit feature code. A palmprint pattern is represented by thousands of ordinal feature codes. In contrast to the state-of-the-art algorithm reported in the literature, our method achieves higher accuracy, with the equal error rate reduced by 42% for a difficult set, while the complexity of feature extraction is halved.

Toward Accurate and Fast Iris Segmentation for Iris Biometrics

Iris segmentation is an essential module in iris recognition because it defines the effective image region used for subsequent processing such as feature extraction. Traditional iris segmentation methods often involve an exhaustive search of a large parameter space, which is time consuming and sensitive to noise. To address these problems, this paper presents a novel algorithm for accurate and fast iris segmentation. After efficient reflection removal, an Adaboost-cascade iris detector is first built to extract a rough position of the iris center. Edge points of iris boundaries are then detected, and an elastic model named pulling and pushing is established. Under this model, the center and radius of the circular iris boundaries are iteratively refined in a way driven by the restoring forces of Hookes law. Furthermore, a smoothing spline-based edge fitting scheme is presented to deal with noncircular iris boundaries. After that, eyelids are localized via edge detection followed by curve fitting. The novelty here is the adoption of a rank filter for noise elimination and a histogram filter for tackling the shape irregularity of eyelids. Finally, eyelashes and shadows are detected via a learned prediction model. This model provides an adaptive threshold for eyelash and shadow detection by analyzing the intensity distributions of different iris regions. Experimental results on three challenging iris image databases demonstrate that the proposed algorithm outperforms state-of-the-art methods in both accuracy and speed.

Ordinal Measures for Iris Recognition

Images of a human iris contain rich texture information useful for identity authentication. A key and still open issue in iris recognition is how best to represent such textural information using a compact set of features (iris features). In this paper, we propose using ordinal measures for iris feature representation with the objective of characterizing qualitative relationships between iris regions rather than precise measurements of iris image structures. Such a representation may lose some image-specific information, but it achieves a good trade-off between distinctiveness and robustness. We show that ordinal measures are intrinsic features of iris patterns and largely invariant to illumination changes. Moreover, compactness and low computational complexity of ordinal measures enable highly efficient iris recognition. Ordinal measures are a general concept useful for image analysis and many variants can be derived for ordinal feature extraction. In this paper, we develop multilobe differential filters to compute ordinal measures with flexible intralobe and interlobe parameters such as location, scale, orientation, and distance. Experimental results on three public iris image databases demonstrate the effectiveness of the proposed ordinal feature models.

Discriminant sparse neighborhood preserving embedding for face recognition

Sparse subspace learning has drawn more and more attentions recently. However, most of the sparse subspace learning methods are unsupervised and unsuitable for classification tasks. In this paper, a new sparse subspace learning algorithm called discriminant sparse neighborhood preserving embedding (DSNPE) is proposed by adding the discriminant information into sparse neighborhood preserving embedding (SNPE). DSNPE not only preserves the sparse reconstructive relationship of SNPE, but also sufficiently utilizes the global discriminant structures from the following two aspects: (1) maximum margin criterion (MMC) is added into the objective function of DSNPE; (2) only the training samples with the same label as the current sample are used to compute the sparse reconstructive relationship. Extensive experiments on three face image datasets (Yale, Extended Yale B and AR) demonstrate the effectiveness of the proposed DSNPE method.

An iris image synthesis method based on PCA and super-resolution

It is very important for the performance evaluation of iris recognition algorithms to construct very large iris databases. However, limited by the real conditions, there are no very large common iris databases now. In this paper, an iris image synthesis method based on principal component analysis (PCA) and super-resolution is proposed. The iris recognition algorithm based on PCA is first introduced and then, iris image synthesis method is presented. The synthesis method first constructs coarse iris images with the given coefficients. Then, synthesized iris images are enhanced using super-resolution. Through controlling the coefficients, we can create many iris images with specified classes. Extensive experiments show that the synthesized iris images have satisfactory cluster and the synthesized iris databases can be very large.

Advances in Biometric Person Authentication

Face.- Texture Features in Facial Image Analysis.- Enhance ASMs Based on AdaBoost-Based Salient Landmarks Localization and Confidence-Constraint Shape Modeling.- Face Authentication Using One-Class Support Vector Machines.- A Novel Illumination Normalization Method for Face Recognition.- Using Score Normalization to Solve the Score Variation Problem in Face Authentication.- Gabor Feature Selection for Face Recognition Using Improved AdaBoost Learning.- An Automatic Method of Building 3D Morphable Face Model.- Procrustes Analysis and Moore-Penrose Inverse Based Classifiers for Face Recognition.- Two Factor Face Authentication Scheme with Cancelable Feature.- Fingerprint.- Local Feature Extraction in Fingerprints by Complex Filtering.- A TSVM-Based Minutiae Matching Approach for Fingerprint Verification.- A Robust Orientation Estimation Algorithm for Low Quality Fingerprints.- An Exact Ridge Matching Algorithm for Fingerprint Verification.- Adaptive Fingerprint Enhancement by Combination of Quality Factor and Quantitative Filters.- Fingerprint Classification Based on Statistical Features and Singular Point Information.- Iris.- An Iterative Algorithm for Fast Iris Detection.- A Non-linear Normalization Model for Iris Recognition.- A New Feature Extraction Method Using the ICA Filters for Iris Recognition System.- Iris Recognition Against Counterfeit Attack Using Gradient Based Fusion of Multi-spectral Images.- An Iris Detection Method Based on Structure Information.- Speaker.- Constructing the Discriminative Kernels Using GMM for Text-Independent Speaker Identification.- Individual Dimension Gaussian Mixture Model for Speaker Identification.- Writing.- Sensor Interoperability and Fusion in Signature Verification: A Case Study Using Tablet PC.- Fusion of Local and Regional Approaches for On-Line Signature Verification.- Text-Independent Writer Identification Based on Fusion of Dynamic and Static Features.- Gait.- Combining Wavelet Velocity Moments and Reflective Symmetry for Gait Recognition.- Model-Based Approaches for Predicting Gait Changes over Time.- Other Biometrics.- Using Ear Biometrics for Personal Recognition.- Biometric Identification System Based on Dental Features.- A Secure Multimodal Biometric Verification Scheme.- Automatic Configuration for a Biometrics-Based Physical Access Control System.

Half-Quadratic-Based Iterative Minimization for Robust Sparse Representation

Robust sparse representation has shown significant potential in solving challenging problems in computer vision such as biometrics and visual surveillance. Although several robust sparse models have been proposed and promising results have been obtained, they are either for error correction or for error detection, and learning a general framework that systematically unifies these two aspects and explores their relation is still an open problem. In this paper, we develop a half-quadratic (HQ) framework to solve the robust sparse representation problem. By defining different kinds of half-quadratic functions, the proposed HQ framework is applicable to performing both error correction and error detection. More specifically, by using the additive form of HQ, we propose an ℓ1-regularized error correction method by iteratively recovering corrupted data from errors incurred by noises and outliers; by using the multiplicative form of HQ, we propose an ℓ1-regularized error detection method by learning from uncorrupted data iteratively. We also show that the ℓ1-regularization solved by soft-thresholding function has a dual relationship to Huber M-estimator, which theoretically guarantees the performance of robust sparse representation in terms of M-estimation. Experiments on robust face recognition under severe occlusion and corruption validate our framework and findings.

Multispectral palm image fusion for accurate contact-free palmprint recognition

In this paper, we propose to improve the verification performance of a contract-free palmprint recognition system by means of feature- level image registration and pixel-level fusion of multi-spectral palm images. Our method involves image acquisition via a dedicated device under contact-free and multi-spectral environment, preprocessing to locate region of interest (ROI) from each individual hand images, feature-level registration to align ROIs from different spectral images in one sequence and fusion to combine images from multiple spectra. The advantages of the proposed method include better hygiene and higher verification performance. Given a database composed of images from 330 hands, two out of four state of the art fusion strategies offer significant performance gain and the best equal error rate (EER) is 0.5%.

Counterfeit iris detection based on texture analysis

This paper addresses the issue of counterfeit iris detection, which is a liveness detection problem in biometrics. Fake iris mentioned here refers to iris wearing color contact lens with textures printed onto them. We propose three measures to detect fake iris: measuring iris edge sharpness, applying Iris-Texton feature for characterizing the visual primitives of iris textures and using selected features based on co-occurrence matrix (CM). Extensive testing is carried out on two datasets containing different types of contact lens with totally 640 fake iris images, which demonstrates that Iris-Texton and CM features are effective and robust in anticounterfeit iris. Detailed comparisons with two state-of-the-art methods are also presented, showing that the proposed iris edge sharpness measure acquires a comparable performance with these two methods, while Iris-Texton and CM features outperform the state-of-the-art.

Robust encoding of local ordinal measures: A general framework of iris recognition

The randomness of iris pattern makes it one of the most reliable biometric traits. On the other hand, the complex iris image structure and various sources of intra-class variations result in the difficulty of iris representation. Although diverse iris recognition methods have been proposed, the fundamentals of iris recognition have not a unified answer. As a breakthrough of this problem, we found that several accurate iris recognition algorithms share a same idea — local ordinal encoding, which is the representation well-suited for iris recognition. After further analysis and summarization, a general framework of iris recognition is formulated in this paper. This work discovered the secret of iris recognition. With the guidance of this framework, a novel iris recognition method based on robust estimating the direction of image gradient vector is developed. Extensive experimental results demonstrate our idea.