This PDF file contains the front matter associated with SPIE Proceedings Volume 6944, including the Title Page, Copyright information, Table of Contents, Introduction, and the Conference Committee listing.

The issues of applying facial recognition at significant distances are non-trivial and often subtle. This paper summarizes 7 years of effort on Face at a distance, which for us is far more than a fad. Our effort started under the DARPA Human Identification at a Distance (HID) program. Of all the programmers under HID, only a few of the efforts demonstrated face recognition at greater than 25ft and only one, lead by Dr. Boult, studied face recognition at distances greater than 50 meters. Two issues were explicitly studied. The first was atmospherics/weather, which can have a measurable impact at these distances. The second area was sensor issues including resolution, field-of-view and dynamic range. This paper starts with a discussion and some of results in sensors related issues including resolution, FOV, dynamic range and lighting normalization. It then discusses the "Photohead" technique developed to analyze the impact of weather/imaging and atmospherics at medium distances. The paper presents experimental results showing the limitations of existing systems at significant distance and under non-ideal weather conditions and presents some reasons for the weak performance. It ends with a discussion of our FASST^TM (failure prediction from similarity surface theory) and RandomEyes^TM approaches, combined into the FIINDER^TM system and how they improved FAAD.

3D face recognition technologies, with a computation time of a few seconds, perform well for person verification. However, current 3D face recognition approaches are too slow for person identification, even for a watch list of only a few hundred face models. By transforming scanned 3D faces into a canonical face format, storage size is greatly compressed and standard feature extraction is enabled: combining these advantages allows a probe scan to be matched to hundreds or thousands of gallery scans in a few seconds on a commodity computer. We report several experiments that extract a sparse feature representation from the canonical 3D face surface and then perform recognition of a probe face based on the sparse features. We expect to have a trade off between algorithm speed and recognition performance. The best results achieved so far are a rank-1 recognition rate of 98.2% and a speed of 1900 face matches per second. Extrapolating these results suggests that multistage systems could achieve comparable or better recognition rates over large galleries within 5 seconds of compute time.

Recently, the iris of the human eye has been used as a biometric indicator for identification. We have witnessed wide-scale deployment of iris technology across many product categories. However, these iris recognition solutions do not reflect the full potential of the technology. The robustness of the standoff iris segmentation approach relies heavily on accurate iris segmentation techniques. Computing iris features requires a high quality segmentation process that focuses on the subject's iris and properly extracts its boundaries. Because iris segmentation is sensitive to the acquisition conditions, it is a very challenging problem. In this paper, we describe a standoff iris recognition system to identify non-cooperative subjects. We introduce a novel iris segmentation approach that takes the analysis of edges into the polar domain at an earlier stage and uses non-iterative polar differential operator to locate the inner and outer borders of the iris. The approach is proven to be very effective for non-ideal gazed and obscured irises while providing comparable results to top performing algorithms on frontal iris images.

Iris image acquisition is the fundamental step of the iris recognition, but capturing high-resolution iris images in real-time is very difficult. The most common systems have small capture volume and demand users to fully cooperate with machines, which has become the bottleneck of iris recognition's application. In this paper, we aim at building an active iris image acquiring system which is self-adaptive to users. Two low resolution cameras are co-located in a pan-tilt-unit (PTU), for face and iris image acquisition respectively. Once the face camera detects face region in real-time video, the system controls the PTU to move towards the eye region and automatically zooms, until the iris camera captures an clear iris image for recognition. Compared with other similar works, our contribution is that we use low-resolution cameras, which can transmit image data much faster and are much cheaper than the high-resolution cameras. In the system, we use Haar-like cascaded feature to detect faces and eyes, linear transformation to predict the iris camera's position, and simple heuristic PTU control method to track eyes. A prototype device has been established, and experiments show that our system can automatically capture high-quality iris image in the range of 0.6m×0.4m×0.4m in average 3 to 5 seconds.

The discriminative capability of a biometric is based on its individuality/uniqueness and is an important factor in choosing a biometric for a large-scale deployment. Individuality studies have been carried out rigorously for only certain biometrics, in particular fingerprint and iris, while work on establishing handwriting and signature individuality has been mainly on feature level. In this study, we present a preliminary individuality model for online signatures using the Fourier domain representation of the signature. Using the normalized Fourier coefficients as global features describing the signature, we derive a formula for the probability of coincidentally matching a given signature. Estimating model parameters from a large database and making certain simplifying assumptions, the probability of two arbitrary signatures to match in 13 of the coefficients is calculated as 4.7x10^-4. When compared with the results of a verification algorithm that parallels the theoretical model, the results show that the theoretical model fits the random forgery test results fairly well. While online signatures are sometimes dismissed as not very secure, our results show that the probability of successfully guessing an online signature is very low. Combined with the fact that signature is a behavioral biometric with adjustable complexity, these results support the use of online signatures for biometric authentication.

Biometric template protection is one of the important issues in deploying a practical biometric system. To tackle this problem, many algorithms have been reported in recent years, most of them being applicable to fingerprint biometric. Since the content and representation of fingerprint template is different from templates of other modalities such as face, the fingerprint template protection algorithms cannot be directly applied to face template. Moreover, we believe that no single template protection method is capable of satisfying the diversity, revocability, security and performance requirements. We propose a three-step cancelable framework which is a hybrid approach for face template protection. This hybrid algorithm is based on the random projection, class distribution preserving transform and hash function. Two publicly available face databases, namely FERET and CMU-PIE, are used for evaluating the template protection scheme. Experimental results show that the proposed method maintains good template discriminability, resulting in good recognition performance. A comparison with the recently developed random multispace quantization (RMQ) biohashing algorithm shows that our method outperforms the RMQ algorithm.

In this paper, we propose a robust spatio-temporal face modelling approach based on multilevel fusion strategy involving cascaded fusion of hybrid multimodal fusion of audio-lip-face motion, correlation and depth features for biometric security application. The proposed approach combines the information from different audio-video based modules, namely: audio-lip motion module, audio-lip correlation module, 2D+3D motion-depth fusion module, and performs a hybrid cascaded fusion in an automatic, unsupervised and adaptive manner, by adapting to the local performance of each module. This is done by taking the output-score based reliability estimates (confidence measures) of each of the module into account. The module weightings are determined automatically such that the reliability measure of the combined scores is maximised. To test the robustness of the proposed approach, the audio and visual speech (mouth) modalities are degraded to emulate various levels of train/test mismatch; employing additive white Gaussian noise for the audio and JPEG compression for the video signals. The results show improved fusion performance for a range of tested levels of audio and video degradation, compared to the individual module performances. Experiments on a 3D stereovision database AVOZES show that, at severe levels of audio and video mismatch, the audio, mouth, 3D face, and tri-module (audio-lip motion, correlation and depth) fusion EERs were 42.9%, 32%, 15%, and 7.3% respectively for biometric identity verification scenario.

Uncontrolled illumination poses severe problems for face recognition in practical application scenarios. Many techniques to deal with this problem rely on illumination modeling and face relighting. In this paper we propose a robust approach to face albedo estimation in the framework of illumination modeling with Spherical Harmonics. This technique requires only a single face image under arbitrary illumination and assumes the face shape is known. The recovered face albedo facilitates face rendering under new illumination conditions which is useful for both illumination invariant face recognition and computer animation. In the proposed approach, the consequences of the violation of the assumption of validity of the spherical harmonics model are mitigated by minimising a cost function involving robust forms of the error in both the spherical harmonics model and the smoothness constraint. The robust estimation provides significantly better results than the traditional Least Squares Estimation in the experiments on a 3D face database.

In this paper, a fast incremental image reduction principal component analysis approach (IIRPCA) is developed for image representation and recognition. As opposed to traditional appearance based image techniques, IRPCA computes the principal components of a sequence of image samples directly on the 2D image matrix incrementally without estimating the covariance matrix. Therefore, IRPCA overcomes the limitations such as the computational cost and memory requirements to making it suitable for real time applications. The feasibility of the proposed approach was tested on a recently published large database consisting of over 2000 face images. IIRPCA shows superiority in terms of computational time, storage and comparable recognition accuracy (94.0%) when compared to recent techniques such as 2DPCA (92.0%) and 2D RPCA (94.5%).

This paper presents a new personal authentication system that simultaneously exploits 2D and 3D palmprint features. Here, we aim to improve the accuracy and robustness of existing palmprint authentication systems using 3D palmprint features. The proposed system uses an active stereo technique, structured light, to capture 3D image or range data of the palm and a registered intensity image simultaneously. The surface curvature based method is employed to extract features from 3D palmprint and Gabor feature based competitive coding scheme is used for 2D representation. We individually analyze these representations and attempt to combine them with score level fusion technique. Our experiments on a database of 108 subjects achieve significant improvement in performance (Equal Error Rate) with the integration of 3D features as compared to the case when 2D palmprint features alone are employed.

We propose a novel cryptographic construct incorporating biometrics which insures a secure communication between two channels just by using Palmprint. The cryptosystem utilizes the advantages of both symmetric and asymmetric cryptographic approaches simultaneously; we denote it as double encryption. Any document in communication is first encrypted using symmetric cryptographic approach; the symmetric key involved is then encrypted using Asymmetric approach. Finally, the concept of fuzzy vault is explored to create a secure vault around the asymmetric key. We investigate the possible usage of palmprints in fuzzy vault to develop a user friendly and reliable crypto system. The experimental results from the proposed approach on the real palmprint images suggest its possible usage in an automated palmprint based key generation system.

This paper presents a new approach to authenticate individuals using triangulation of hand vein images. The proposed method is fully automated and employs palm dorsal hand vein images acquired from the low-cost, near infrared, contactless imaging. The knuckle tips are used as key points for image normalization and the extraction of region of interest. The matching scores are generated in two parallel stages; (i) hierarchical matching score from the four topologies of triangulation in binarized vein structures and (ii) from the geometrical features consisting of knuckle point perimeter distances in the acquired images. The weighted score level combination from these two matching scores are used to authenticate the individuals. The achieved experimental results from the proposed system using contactless, palm dorsal hand vein images are promising and suggest more user friendly alternative for user identification.

Recent investigations indicate cardiovascular function is a viable biometric. This paper explores biometric techniques based on multiple modalities for sensing cardiovascular function. Analysis of data acquired with an electrocardiogram (ECG) combined with corresponding data from pulse oximetry and blood pressure indicates that features can be extracted from the signals, which correspond to individuals. While a person's heart rate can vary with mental and emotional state, certain features corresponding to the heartbeat appear to be unique to the individual. Our protocol induced a range of mental and emotional states in the subject and the analysis identifies features of the cardiovascular signals that are invariant to mental and emotional state. Furthermore, the three measures of cardiovascular function provide independent information, which can be fused to achieve robust performance compared to a single modality.

As biometrics gains popularity, there is an increasing concern about privacy and misuse of biometric data held in central repositories. Furthermore, biometric verification systems face challenges arising from noise and intra-class variations. To tackle both problems, a multimodal biometric verification system combining fingerprint and voice modalities is proposed. The system combines the two modalities at the template level, using multibiometric templates. The fusion of fingerprint and voice data successfully diminishes privacy concerns by hiding the minutiae points from the fingerprint, among the artificial points generated by the features obtained from the spoken utterance of the speaker. Equal error rates are observed to be under 2% for the system where 600 utterances from 30 people have been processed and fused with a database of 400 fingerprints from 200 individuals. Accuracy is increased compared to the previous results for voice verification over the same speaker database.

Multimodal biometric systems allow to overcome some of the problems presented in unimodal systems, such as non-universality, lack of distinctiveness of the unimodal trait, noise in the acquired data, etc. Integration at the matching score level is the most common approach used due to the ease in combining the scores generated by different unimodal systems. Unfortunately, scores usually lie in application-dependent domains. In this work, we use linear logistic regression fusion, in which fused scores tend to be calibrated log-likelihood-ratios and thus, independent of the application. We use for our experiments the development set of scores of the DS2 Evaluation (Access Control Scenario) of the BioSecure Multimodal Evaluation Campaign, whose objective is to compare the performance of fusion algorithms when query biometric signals are originated from heterogeneous biometric devices. We compare a fusion scheme that uses linear logistic regression with a set of simple fusion rules. It is observed that the proposed fusion scheme outperforms all the simple fusion rules, with the additional advantage of the application-independent nature of the resulting fused scores.

Fingerprints are being extensively used for person identification in a number of commercial, civil, and forensic applications. Most of the current fingerprint verification systems utilize features that are based on minutiae points and ridge patterns. While minutiae based fingerprint verification systems have shown fairly high accuracies, further improvements in their performance are needed for acceptable performance, especially in applications involving very large scale databases. In an effort to extend the existing technology for fingerprint verification, we propose a new representation and matching scheme for fingerprint using Scale Invariant Feature Transformation (SIFT). We extract characteristic SIFT feature points in scale space and perform matching based on the texture information around the feature points using the SIFT operator. A systematic strategy of applying SIFT to fingerprint images is proposed. Using a public domain fingerprint database (FVC 2002), we demonstrate that the proposed approach complements the minutiae based fingerprint representation. Further, the combination of SIFT and conventional minutiae based system achieves significantly better performance than either of the individual schemes.

Singular points are critical features of fingerprints. They are broadly used in fingerprint registration and many other applications. The rotation estimation of singular points can largely affect the performance of a fingerprint registration method. Most existing algorithms detect singular points directly from the available ridge orientation structures. However, the performance degrades if the singular patterns in a fingerprint image are incomplete. In this paper, we propose a model-based method for estimating the angular difference between singular points. The proposed method exploits analytical features derived from ridge orientation models and is more robust with incomplete fingerprints. We test the proposed method with manually rotated fingerprints generated from the FVC2002 DB1a database. The performance is evaluated by mean square errors (MSE) and ROC curves.

Fingerprint recognition is one of the popular used methods of biometrics. However, due to the surface topography limitation, fingerprint recognition scanners are easily been spoofed, e.g. using artificial fingerprint dummies. Thus, biometric fingerprint identification devices need to be more accurate and secure to deal with different fraudulent methods including dummy fingerprints. Previously, we demonstrated that Optical Coherence Tomography (OCT) images revealed the presence of the artificial fingerprints (made from different household materials, such as cement and liquid silicone rubber) at all times, while the artificial fingerprints easily spoofed the commercial fingerprint reader. Also we demonstrated that an analysis of the autocorrelation of the OCT images could be used in automatic recognition systems. Here, we exploited the three-dimensional (3D) imaging of the artificial fingerprint by OCT to generate vivid 3D image for both the artificial fingerprint layer and the real fingerprint layer beneath. With the reconstructed 3D image, it could not only point out whether there exists an artificial material, which is intended to spoof the scanner, above the real finger, but also could provide the hacker's fingerprint. The results of these studies suggested that Optical Coherence Tomography could be a powerful real-time noninvasive method for accurate identification of artificial fingerprints real fingerprints as well.

For the domain of strategy-based behavioral biometrics we propose the concept of profiles enhanced with spatial, temporal and contextual information. Inclusion of such information leads to a more stable baseline profile and as a result more secure systems. Such enhanced data is not always readily available and often is time consuming and expensive to acquire. One solution to this problem is the use of artificially generated data. In this paper a novel methodology for creation of feature-level synthetic biometric data is presented. Specifically generation of behavioral biometric data represented by game playing strategies is demonstrated. Data validation methods are described and encouraging results are obtained with possibility of expanding proposed methodologies to generation of artificial data in the domains other then behavioral biometrics.

Statistical modeling of biometric systems at the score level is extremely important. It is the foundation of the performance assessment of biometric systems including determination of confidence intervals and test sample size for simulations, and performance prediction of real world systems. Statistical modeling of multimodal biometric systems allows the development of a methodology to integrate information from multiple biometric sources. We present a novel approach for estimating the marginal biometric matching score distributions by using extreme value theory in conjunction with non-parametric methods. Extreme Value Theory (EVT) is based on the modeling of extreme events represented by data which has abnormally low or high values in the tails of the distributions. Our motivation stems from the observation that the tails of the biometric score distributions are often difficult to estimate using other methods due to lack of sufficient numbers of training samples. However, good estimates of the tails of biometric distributions are essential for defining the decision boundaries. We present EVT based novel procedures for fitting a score distribution curve. A general non-parametric method is used for fitting the majority part of the distribution curve, and a parametric EVT model - the general Pareto distribution - is used for fitting the tails of the curve. We also demonstrate the advantage of applying the EVT by experiments.

In this paper, a novel micro fingerprint sensor was proposed using densely arrayed micro heaters as the sensing elements. Because it senses the temperature differences between the touching ridges and non-touching valleys of fingerprint patterns, there is no electrical static discharge (ESD), and promising to realize robust sensing. By MEMS process, it could be fabricated portably small size. Silicon-based and poyimide (PI) film-based approach was tried. In the first case, due to its great thermal conductivity (λ-silicon: 150 W/m•K), special structure was necessary for the thermal isolation that make the fabrication process complex and expensive. In the latter case, with the thickness of 25μm PI film (λ: 0.12 W/m•K) as the substrate, other special structure was not necessary for further thermal isolation, and flexible enough to be easily attached to non-planar surface. The whole fabrication process was simple and achieved in a low temperature range (< 130°C). The arrayed sensing elements were micro resistors made from sputtered platinum. Because of its small thermal capacity (sensing element: 200nm thick /10μm wide) and effective thermal isolation, this sensor was very sensitive (at 5V input, ΔT: 270°C /0.1μs). Characteristics of the sensor (silicon and PI film-based) were introduced, and the sensing principle was demonstrated by experiments.

3D facial feature point localization is very important to registration. This paper proposes a localization method that is capable of locating 3D facial feature points rapidly while achieving high localization and registration accuracy. There are two contributions of this paper. The first is the introduction of the Cascade PCA which allows the non-occluded and symmetric face models to be normalized quickly while spending more computation on occluded face models. The second is the three face shape models which are used to verify the normalization results produced by Cascade PCA, and localize dozens of feature points at the same time. Experimental results prove the efficiency and accuracy of our method both in localization and registration.

Learning curve phenomenon indicates that not all available images need to be used in training. This paper proposes a three-step intelligent sampling to construct a representative and efficient training database, where both the number of training images and which images to be included are determined. Firstly, clustering on a subset of huge face database is implemented as preparation. Secondly, systematic sampling on clusters is utilized to improve the efficiency. Thirdly, performance is evaluated to check whether the learning curve has reached a point of diminishing returns, and a new metric of difficulty is defined to determine which images from the complementary subset of initial training set should be added into training. The proposed intelligent three step sampling design enhances recognition rate and generalizability while improving efficiency, which exerts the full potential of any given face recognition algorithm without system overhaul.

Many methods based on biometrics such as fingerprint, face, iris, and retina have been proposed for person identification. However, for deceased individuals, such biometric measurements are not available. In such cases, parts of the human skeleton can be used for identification, such as dental records, thorax, vertebrae, shoulder, and frontal sinus. It has been established in prior investigations that the radiographic pattern of frontal sinus is highly variable and unique for every individual. This has stimulated the proposition of measurements of the frontal sinus pattern, obtained from x-ray films, for skeletal identification. This paper presents a frontal sinus recognition method for human identification based on Image Foresting Transform and shape context. Experimental results (ERR = 5,82%) have shown the effectiveness of the proposed method.

The iris contains fibrous structures of various sizes and orientations which can be used for human identification. Drawing from a directional energy iris identification technique, this paper investigates the size, orientation, and location of the iris structures that hold stable discriminatory information. Template height, template width, filter size, and the number of filter orientations were investigated for their individual and combined impact on identification accuracy. Further, the iris was segmented into annuli and radial sectors to determine in which portions of the iris the best discriminatory information is found. Over 2 billion template comparisons were performed to produce this analysis.

Rather than use arbitrary matching threshold values and a heuristic set of features while comparing minutiae points during the fingerprint verification process, we develop a system which considers only the optimal features, which contain the highest discriminative power, from a predefined feature set. For this, we use a feature selection algorithm which adds features, one at a time, till it arrives at an optimal feature set of the target size. The classifier is trained on this feature set, on a two class problem representing pairs of matched minutiae points belonging to fingerprints of same and different users. During the test phase, the system generates a number of candidate matched minutiae pairs; features from each of them are extracted and given to the classifier. Those that are incorrectly matched are eliminated from the scoring algorithm. We have developed a set of seven candidate features, and tested our system using the FVC 2002 DB1 fingerprint database. We study how feature sets of different sizes affect the accuracy of the system, and observe how additional features not necessarily would improve the performance of a classifier. This is illustrated in how using a 3 feature set gives us the most accurate system and using bigger feature sets cause a slight drop in accuracy.

In this paper, human motion model and RCS (radar cross section) simulation of radar returns from human are investigated. Micro-Doppler signatures [1-6] induced by human motions are studied. It shows that the time-frequency representation of micro-Doppler signature provides distinctive time-varying features for human motions. Motion of different body part has different micro-Doppler signature. Thus, micro-Doppler can be a promising method for classifying human activities. Measurement data using an experimental X-band micro-Doppler radar were collected and the results are compared with the corresponding simulation results. The classification of human motions based on micro-Doppler signatures is also discussed.