Search

Although several automatic computer systems have been proposed to address facial expression recognition problems, the majority of them still fail to cope with some requirements of many practical application scenarios. In this paper, one of the most influential and common issues raised in practical application scenarios when applying automatic facial expression recognition system, head pose variation, is comprehensively explored and investigated. In order to do this, two novel texture feature representations are proposed for implementing multi-view facial expression recognition systems in practical environments. These representations combine the block-based techniques with Local Ternary Pattern-based features, providing a more informative and efficient feature representation of the facial images. In addition, an in-house multi-view facial expression database has been designed and collected to allow us to conduct a detailed research study of the effect of out-of-plane pose angles on the performance of a multi-view facial expression recognition system. Along with the proposed in-house dataset, the proposed system is tested on two well-known facial expression databases, CK+ and BU-3DFE datasets. The obtained results shows that the proposed system outperforms current state-of-the-art 2D facial expression systems in the presence of pose variations.

A new multimodal biometric database designed and acquired within the framework of the European BioSecure Network of Excellence is presented. It is comprised of more than 600 individuals acquired simultaneously in three scenarios: 1) over the Internet, 2) in an office environment with desktop PC, and 3) in indoor/outdoor environments with mobile portable hardware. The three scenarios include a common part of audio/video data. Also, signature and fingerprint data have been acquired both with desktop PC and mobile portable hardware. Additionally, hand and iris data were acquired in the second scenario using desktop PC. Acquisition has been conducted by 11 European institutions. Additional features of the BioSecure Multimodal Database (BMDB) are: two acquisition sessions, several sensors in certain modalities, balanced gender and age distributions, multimodal realistic scenarios with simple and quick tasks per modality, cross-European diversity, availability of demographic data, and compatibility with other multimodal databases. The novel acquisition conditions of the BMDB allow us to perform new challenging research and evaluation of either monomodal or multimodal biometric systems, as in the recent BioSecure Multimodal Evaluation campaign. A description of this campaign including baseline results of individual modalities from the new database is also given. The database is expected to be available for research purposes through the BioSecure Association during 2008, Published at IEEE Transactions on Pattern Analysis and Machine Intelligence journal

This paper investigates the effect of earthquake ground motion duration on the design and collapse risk of reinforced concrete shear wall buildings from 6 to 30 stories. Both design and collapse levels of shaking were considered through nonlinear incremental dynamic analysis. At the design level of shaking, it was found that (1) maximum interstory drifts were increased as ground motion duration increased (from

Recent post-earthquake evaluations of the performance of light-frame wood buildings have pointed out a number of deficiencies that may make such buildings susceptible to high levels of damage and collapse in a seismic event. This chapter focuses, firstly, on defining the deficiencies that are common in light-frame wood structures, including weak first storey, weak roof or floor diaphragms, shear walls with insufficient strength, inadequate load path, geometric and mass eccentricities, and brittle components. Subsequently, the chapter describes conventional as well as state-of-the-art retrofit solutions that are available in specific codes and guidelines.

© 2019 by the authors Red algae of the genus Plocamium have been a rich source of halogenated monoterpenes. Herein, a new cyclic monoterpene, costatone C (7), was isolated from the extract of P. angustum collected in New Zealand, along with the previously reported (1E,5Z)-1,6-dichloro-2-methylhepta-1,5-dien-3-ol (8). Elucidation of the planar structure of 7 was achieved through conventional NMR and (−)-HR-APCI-MS techniques, and the absolute configuration by comparison of experimental and DFT-calculated ECD spectra. The absolute configuration of 8 was determined using Mosher’s method. Compound 7 showed mild antibacterial activity against Staphylococcus aureus and S. epidermidis. The state of Plocamium taxonomy and its implications upon natural product distributions, especially across samples from specimens collected in different countries, is also discussed.

In the context of facial expression recognition (FER), this paper reviews the fundamental theories of emotions and further explains the key dimensions of a defined emotional space. The main contribution of this paper is to propose a set of novel categorization methods for facial expressions to be used in the design of an automatic FER system. This novel categorization enables the facial expression to be interpreted in a better way that and to be more effective in practical applications of automatic FER systems. In order to validate the feasibility of the proposed categorization methods, a set of experiments is reported which investigates and analyzes the influence that the novel categorization brings to a multi-view FER system.

Seismic reliability analyses account for the inherent uncertainties in both the actions (earthquakes) and the reactions (properties of the structural systems) of a structure. To predict the failure probability of a structure, the system response due to external loads is usually estimated by a numerical method. In this paper, seismic reliability analyses were performed on a novel timber-steel hybrid system labelled FFTT (Finding the Forest Through the Trees) system. The FFTT system utilizes mass-timber panels to resist gravity and lateral loads and interconnecting steel members to provide the necessary ductility for seismic demands. To reduce the computational effort for reliability analyses, Genetic Algorithms (GA) and Analysis of Variance in combination with response surface methods were applied and compared. Uncertainties involving ground motions, seismic weight, connection properties of the lateral load resisting system, and ductility factor were considered in formulating the performance functions. Mean and standard deviation of peak inter-storey drift were selected as performance criteria. Nonlinear dynamic analyses were run to generate the response database for the FFTT system and the reliability index was calculated using second-order reliability methods. The results showed that the GA method was superior and that the ground motion was the most significant factor for structural reliability, while the ductility factor, the structural weight, the hold-down and connection stiffness also played significant roles.

Interest in the exploitation of soft biometrics information has continued to develop over the last decade or so. In comparison with traditional biometrics, which focuses principally on person identification, the idea of soft biometrics processing is to study the utilisation of more general information regarding a system user, which is not necessarily unique. There are increasing indications that this type of data will have great value in providing complementary information for user authentication. However, the authors have also seen a growing interest in broadening the predictive capabilities of biometric data, encompassing both easily definable characteristics such as subject age and, most recently, `higher level' characteristics such as emotional or mental states. This study will present a selective review of the predictive capabilities, in the widest sense, of biometric data processing, providing an analysis of the key issues still adequately to be addressed if this concept of predictive biometrics is to be fully exploited in the future.

An off-line text-independent writer verification system that leverages the similarities with the field of speaker recognition by employing analogous techniques for modelling and comparing the features extracted from the input text images is presented. The main contribution of this work is the use of the i -vector paradigm in a writer verification setting. The proposed system is evaluated with images of lines of text from the IAM Handwriting Database, and compared with more traditional approaches. The authors also analyse several algorithms for the detection and extraction of points of interest in the text images, different parameters for the modelling part and different scoring techniques. The obtained results show that the use of i -vectors clearly improves the performance of the system even for configurations where the overhead of the additional calculations is minimal.

© 2019 IEEE. The use of soft-biometric data as an auxiliary tool on user identification is already well known. Gender, handorientation and emotional state are some examples which can be called soft-biometrics. These soft-biometric data can be predicted directly from the biometric templates. It is very common to find researches using physiological modalities for soft-biometric prediction, but behavioural biometric is often not well explored for this context. Among the behavioural biometric modalities, keystroke dynamics and handwriting signature have been widely explored for user identification, including some soft-biometric predictions. However, in these modalities, the soft-biometric prediction is usually done in an individual way. In order to fill this space, this study aims to investigate whether the combination of those two biometric modalities can impact the performance of a soft-biometric data, gender prediction. The main aim is to assess the impact of combining data from two different biometric sources in gender prediction. Our findings indicated gains in terms of performance for gender prediction when combining these two biometric modalities, when compared to the individual ones.

“Natural revocability” is an extremely simple and intuitive strategy for the revocation process in the event of compromise without the need for complex mathematical processing, This paper investigates the possibilities of adopting the strategy, in relation to security and reliability in handwritten signature analysis by exploring the effect of writer style, age and gender on signature analysis. This is examined by performing analysis of variance for the extracted features based on three style categories, two gender groups and three age groups. The results from the analysis provide some valuable insight into the concept of natural revocability as a function of writer style, age and gender.

The damage caused by large subduction earthquakes is due in part to high number of load reversal cycles. Experimental and analytical studies indicate that shaking duration and number of cycles contribute to the damage. Currently, building codes do not include explicit design provisions for shaking duration. This paper investigates how shaking duration affects the response of tall, shearwall buildings in British Columbia, which is located in the Cascadia Subduction Zone. A suite of concrete shearwall archetype building models are analyzed with suites long and short duration ground motions. A case study of a reinforced concrete frame is presented to illustrate that long duration shaking can also affect significantly their seismic response. The results are useful for the elaboration of design provisions to account for shaking duration.

Facial expressions can be seen as a form of non-verbal communication as well as a primary means of conveying social information among humans.Automatic facial expression recognition (FER) can be applied to a wide range of scenarios in human-computer interaction, facial animation, entertainment, and psychology studies. For feature representation in a FER system, various texture descriptors have been employed to derive an effective solution for this system. However, these individual texture descriptor-based FER systems have often failed to achieve effective performance in the recognition of facial expressions. In this sense, it is necessary to further improve the general performance of a facial expression recognition system, evaluating different feature representations. In this paper, a novel local descriptor for a facial expression recognition system is proposed, designated the level of difference descriptor (LOD). The main goal is to use this descriptor as a supplement to state-of-the-art local descriptors to further improve the performance of a FER system in terms of classification accuracy. Furthermore, the fusion of various texture features for devising a robust feature representation for multi-view facial expression recognition is presented.

Biometrics is the scientific discipline of identifying individuals by means of the measurement of unique personal attributes, such as facial appearance, fingerprints, iris patterning, the voice, the way we write, and the way we walk. The new technologies of biometrics have a wide range of practical applications, from securing mobile phones and laptops to establishing identity in bank transactions, travel documents, and national identity cards. Biometrics: A Very Short Introduction considers the capabilities of biometrics-based identity checking and looks at the basic techniques in use today, developments in system design, and emerging technologies. It also explores the benefits and limitations of biometrics technologies, and how they can effectively support our increasingly interconnected society.

In order to assess the practicality or suitability of a particular biometric modality in any given application, or to judge how effective biometrics-based identity monitoring is likely to be in a particular situation, we need to understand the principal issues that affect a chosen modality. ‘Making biometrics work’ examines the relevant technologies of four major modalities—fingerprints, iris patterns, facial images, and the handwritten signature. All modalities offer both advantages and disadvantages, and performance will depend very much on the nature of the proposed application, the nature of the population who will be the primary users of the system, and the environment in which it is to operate.

‘Where are we going?’ looks at some other biometric modalities that may achieve a higher profile in future, including hand-based biometrics, keystroke dynamics, ear-shape biometrics, and electrocardiogram and electroencephalogram measurements. It then discusses the problem of ageing in the design and use of biometric systems before considering human interaction with biometric systems and the impact of the usability of systems on their success. Biometrics systems are increasingly and manifestly having an impact on everyday life. Properly understood and sensitively applied, these can improve our security, increase our confidence in processes that are designed to dovetail into the way modern society works, and enhance the convenience and power of technology, for the benefit of everyone.

Many aspects of modern life raise questions about individual identity: how we prove that we are who we say we are, and how we can be sure that another person is who he or she claims to be. ‘Are you who you say you are?’ considers the different ways of establishing personal identity and the fundamentals of biometrics. In principle, any human characteristic can be used as a biometric data source provided it meets the following four basic criteria: universality, uniqueness, permanence, and collectability. The source of the measurement data used for identification—biometric modality—can be either physiological or behavioural. Some applications of biometric systems are also considered.

‘Enhancing biometric processing’ explores how the field of biometrics is developing, and the main ideas promoting the improvement of the accuracy, reliability, and effectiveness of biometric systems. It is unlikely that any individual biometric modality operating alone will completely meet all the desirable criteria for a given task, especially when the variety of issues that are needed are considered in any practical situation. To improve system performance different systems can be used such as adding extra power using a multiclassifier configuration, increasing flexibility using multimodal systems, and using soft biometrics as additional identity evidence. Resistance to ‘spoofing’ attacks, biometric data integrity, and extending the application domains for biometrics-based processing are also considered.

Is there any reason why the data captured in a biometric system could not be used for other sorts of prediction too? ‘An introduction to predictive biometrics’ looks at how instead of supplementing conventional data with soft biometric data, conventional biometric data might be used to predict some soft biometric characteristics of an individual. Predictive biometrics has expanded considerably in recent years, with the two most frequently studied characteristics being age and gender using, for example, biometric iris data. Recent studies are taking predictive biometrics further still, aiming to predict so-called ‘higher level’ individual characteristics, such as those which reflect an individual’s mental or emotional state.

‘Biometrics: where should I start?’ considers what constitutes a practical biometric system. It looks at the principles on which such a system operates, builds up a picture of the components needed to construct such a system, and takes the first steps towards understanding how to implement a biometric system. It also looks at how, when, and why errors can arise and how the performance of a biometric system can be evaluated, both qualitatively and quantitatively, to understand more clearly the nature of the interaction between the user and the system itself, and to determine helpful ways of describing both basic system factors and user characteristics, which will ultimately influence system performance.

Summary Performance-based design offers a more direct, non-prescriptive and rational approach over more traditional approaches used for the design of buildings and other structures. However, performance-based design requires the use of extensive nonlinear analyses on three-dimensional building models and typically requires significant computational capabilities and/or time to conduct such analyses. A practical way to overcome these limitations is to utilize recent advantages in parallel computing using cloud-based servers to conduct the necessary analyses. This approach can be used as a cost-effective way to conduct structural analyses in a fraction of the time compared with traditional computational methods. This paper explores the use of parallel and cloud computing in the performance-based design and analysis of tall buildings. Two case studies are presented that highlight the application of high-performance computing for the nonlinear dynamic analysis of a detailed 52-story building model. These case studies highlight both the cost and time benefits provided by high-performance computing for performance-based earthquake engineering. Copyright © 2015 John Wiley & Sons, Ltd.

Handwriting biometrics has a long history, especially when the handwritten signature is the target, but it has also proved possible to use handwriting as a basis for the prediction of various non-unique but forensically useful characteristics of the writer, generally considered to be examples of so-called ‘soft biometrics’. Most commonly, these are characteristics such as the age or gender of the writer, but the predictive capabilities arising in handwriting offer wider opportunities for trait prediction. This study presents a preliminary investigation of the use of handwriting to predict information about the writer relating specifically to higher level characteristics such as emotional state. The authors present an initial study to demonstrate that this is possible, and explore a number of factors particularly relevant to the use of such a capability in areas of forensic investigation.

As biometric systems are deployed in increasingly diverse applications, it becomes correspondingly important to understand the impact which human aging has on system performance. Aging directly affects those physiological and behavioral traits which are characterized in biometric measurements, and a practical biometric system must be designed to account for age-induced changes. However, age can also have very positive implications, for example as a source of further identification information. This paper reviews research to understand how age factors impinge on biometric systems and uses this to synthesize a system infrastructure to unify implementation principles. We present new results to show how multiagent structures can provide an effective framework for this purpose, enhancing performance in both identification and predictive scenarios.

Iris-based biometric technologies have been widely adopted in recent years and implemented in many different scenarios. This has meant that increasing flexibility is now being sought in terms of how systems are configured to improve user-system interaction, while also producing a developing opportunity for applications in forensic analysis. Current typical iris biometric deployments, while generally expected to perform well, require a considerable level of co-operation from the system user. Specifically, the physical positioning of the human eye in relation to the iris capture device is a critical factor, which can substantially affect the performance of the overall iris biometric system. The work reported in this study will explore the acquisition of iris images captured at varying positions with respect to the capture device, and in particular presents a preliminary investigation of types of iris images captured when the gaze angle of a subject is not aligned with the axis of the camera lens.

As biometrics-based identity authentication systems have become more widely deployed, it has become evident that traditional identification and verification tasks are not the only application for such approaches. The prediction of individual, but non-unique, characteristics such as subject age is also an obvious option, since there are diverse situations in which information short of absolute identity is itself valuable. Physical ageing is an important issue for practical biometrics, since it is known that the associated physiological changes can impair performance for most modalities. Understanding the effects of ageing is necessary, therefore, both to optimise attainable performance but also to understand how to manage biometric templates, especially as the time elapsed between enrolment and use increases. Age prediction is relatively poorly represented in the literature. This chapter will explore applications of age prediction from iris biometrics and the implications for the underpinning computational structures.

Clustering is deemed one of the most difficult and challenging problems in machine learning. In this paper, we propose a multilocal search and adaptive niching-based genetic algorithm with a consensus criterion for automatic data clustering. The proposed algorithm employs three local searches of different features in a sophisticated manner to efficiently exploit the decision space. Furthermore, we develop an adaptive niching method, which can dynamically adjust its parameter value depending on the problem instance as well as the search progress, and incorporate it into the proposed algorithm. The adaptation strategy is based on a newly devised population diversity index, which can be used to promote both genetic diversity and fitness. Consequently, diverged niches of high fitness can be formed and maintained in the population, making the approach well-suited to effective exploration of the complex decision space of clustering problems. The resulting algorithm has been used to optimize a consensus clustering criterion, which is suggested with the purpose of achieving reliable solutions. To evaluate the proposed algorithm, we have conducted a series of experiments on both synthetic and real data and compared it with other reported methods. The results show that our proposed algorithm can achieve superior performance, outperforming related methods.

Cancellable biometrics adopt an approach in which biometric templates can be cancelled and replaced if compromised, for example if they are lost or stolen, and can therefore overcome some of the security concerns about biometric-based authentication systems. This paper proposes a simple and effective template protection method (cancellable transformation) for providing cancellable data, called the double sum (DS) method, which performs a sum procedure over the attributes that makes the definition of the original data a hard (and almost impossible) process. In order to investigate the feasibility of the proposed method, an empirical analysis is conducted and we use as examples two different biometric modalities (face and voice) separately and in a multi-modal context (multi-biometric). The main aim of this paper is to provide greater security and improved performance in the biometric authentication process. The datasets used in this analysis were TIMIT for voice and the AR Face dataset for face. As a result of this analysis, we will observe that the proposed transformation offered higher performance to two existing cancellable transformations.

The concept of cancellable biometrics has been introduced as a way to overcome privacy concerns surrounding the management of biometric data. The goal is to transform a biometric trait into a new but revocable representation for enrolment and identification/verification. Thus, if compromised, a new representation of original biometric data can be generated. In addition, multi-biometric systems are increasingly being deployed in various biometric-based applications because of their advantages over uni-biometric systems. In this study, the authors specifically investigate the use of ensemble systems and cancellable transformations for the multi-biometric context, and the authors use as examples two different biometric modalities (fingerprint and handwritten signature) separately and in the multi-modal context (multi-biometric). The datasets to be used in this analysis were FVC2004 (fingerprint verification competition) for fingerprint and an in-house database for signature. To increase the effectiveness of the proposed ensemble systems, two feature selection (FS) methods will be used to distribute the attributes among the individual classifiers of an ensemble, increasing diversity and performance of such systems. As a result of this analysis, they will observe that the use of a cancellable transformation in the multi-biometric dataset increased accuracy level for the ensemble systems, mainly when using FS methods.

In this paper, a generic framework for historical manuscript image and data processing, visualisation and analysis is introduced with a focus on the modelling of manuscript metadata underpinning the interaction. The goal of such a framework is to capture the requirements from three types of activities involving historical manuscripts: presentation, management and analysis. In addition to an overall text-based description of an historical manuscript, a central requirement of such a framework is to associate rich media information (e.g. video, flash component, etc.) to the manuscript or a specific region of the manuscript. A second requirement is to enable interchange of the manuscript data as well as the attached information between users. As a result of an extensive analysis of requirements collected across a wide range of target user groups, an XML-based metadata language derived from a relational database model is proposed to form an historical document data model, and a prototype system is developed to demonstrate some of the advanced functionalities enabled by this data model. Thus, the proposed framework provides an important tool in promoting access to historical documents on a wide and diverse basis, embracing the fundamental principles of universal access to a shared cultural heritage.

This paper introduces an empirical investigation which directly addresses and explores gender prediction capacity from digitised handwriting data from several different perspectives - such as feature type (static/dynamic) and content (fixed/variable) types - in order to provide extensive experimental evidence and analysis to guide the development of a better understanding of the opportunities for and practical consequences of gender prediction from digital handwriting data.

In current force-based seismic design procedures, structures are allowed to behave inelastically during significant seismic events. For this reason, the applied design base shear of a structure is calculated by reducing the elastic strength demand by a ductility factor that represents the ability of a structure or structural system to deform inelastically beyond yielding. Whereas North American building codes provide ductility factors for traditional structural systems, including timber-based or steel-based systems, there are currently no design provisions available for novel hybrid structural systems. Thus research is required to define the force reduction factors, and specifically the ductility factor, Rd, to safely and efficiently design novel systems within the existing seismic design provisions. One such novel system is the so-called finding the forest through the trees (FFTT), proposed in 2012, where mass-timber panels act as shear walls and are connected to each other and to perimeter frame...

The analysis of handwritten signatures is an area of ongoing and increasing interest for both biometrics and forensics researchers. Hesitation in writing execution, estimated (to a large degree) through human inspection, has been commonly used by Forensic Document Examiners (FDE) or Forensic Handwriting Experts (FHE) as one of the important features which may be used to distinguish between genuine signatures and imitations or forgeries. This paper introduces some simple objective measures of hesitation and investigates their reliability and efficacy in automating the analysis of handwritten signatures for forensic or biometric applications.

Recent technological advances have enhanced the computer-based indexing and searching of digitised printed books. The performance now achievable in this domain, however, does not at present extend to handwritten texts which inherently contain more significant letter-based variation within their content. Furthermore, in most studies that address the handwritten text retrieval problem, a large training dataset is required which, very often, influences the context and search lexicon. In this paper a novel method is described to overcome the training data problem using a character-based modelling (termed grapheme spectrum) approach and a word modelling technique (termed synthesised word) enabling the retrieval of keywords that have not explicitly been seen in the training set. When tested on an illustrative historical manuscript the performance of the proposed word retrieval technique shows a clear advantage over existing methods.

PurposeBiometric authentication, which requires storage of biometric templates and/or encryption keys, raises a matter of serious concern, since the compromise of templates or keys necessarily compromises the information secured by those keys. To address such concerns, efforts based on dynamic key generation directly from the biometrics have recently emerged. However, previous methods often have quite unacceptable authentication performance and/or small key spaces and therefore are not viable in practice. The purpose of this paper is to propose a novel method which can reliably generate long keys while requires storage of neither biometric templates nor encryption keys.Design/methodology/approachThis proposition is achieved by devising the use of fingerprint orientation fields for key generation. Additionally, the keys produced are not permanently linked to the orientation fields, hence, allowing them to be replaced in the event of key compromise.FindingsThe evaluation demonstrates that the proposed method for dynamic key generation can offer both good reliability and security in practice, and outperforms other related methods.Originality/valueIn this paper, the authors propose a novel method which can reliably generate long keys while requires storage of neither biometric templates nor encryption keys. This is achieved by devising the use of fingerprint orientation fields for key generation. Additionally, the keys produced are not permanently linked to the orientation fields, hence, allowing them to be replaced in the event of key compromise.

The effectiveness with which individual identity can be predicted in, for example, an antiterrorist scenario can benefit from seeking a broad base of identity evidence. The issue of improving performance can be addressed in a number of ways, but system configurations based on integrating different information sources (often involving more than one biometric modality) are a widely adopted means of achieving this. This paper presents a new approach to improving identification performance, where both direct biometric samples and “soft-biometric” knowledge are combined. Specifically, however, we propose a strategy based on an intelligent agent-based decision-making process, which predicts both absolute identity and also other individual characteristics from biometric samples, as a basis for a more refined and enhanced overall identification decision based on flexible negotiation among class-related agents.

The main aim of biometric-based identification systems is to automatically recognize individuals based on their physiological and/or behavioural characteristics such as fingerprint, face, hand-geometry, among others. These systems offer several advantages over traditional forms of identity protection. However, there are still some important aspects that need to be addressed in these systems. The main questions are concerned with the security of biometric authentication systems since it is important to ensure the integrity and public acceptance of these systems. In order to avoid the problems arising from compromised biometric templates, the concept of cancellable biometrics has recently been introduced. The concept is to transform a biometric trait into a new representation for enrolment and matching. Although cancellable biometrics were proposed to solve privacy concerns, the concept raises new issues, since they make the authentication problem more complex and difficult to solve. Thus, more effective authentication structures are needed to perform these tasks. In this paper, we investigate the use of ensemble systems in cancellable biometrics, using fingerprint-based identification to illustrate the possible benefits accruing. In order to increase the effectiveness of the proposed ensemble systems, three feature selection methods will be used to distribute the attributes among the individual classifiers of an ensemble. The main aim of this paper is to analyse the performance of such well-established structures on transformed biometric data to determine whether they have a positive effect on the performance of this complex and difficult task.

In this paper, we present a novel and robust road tracking system for vision-based personal navigation. Novelty of the work includes the use of multiple Condensation filters to track the road of arbitrary shape and automatic switching between trackers according to road conditions. The approach allows the road to be represented as a simple hyperbola. It also supports the representation of the road as a sequence of connected arcs/segments so that information from a digital map can be integrated into tracking. The parameters of the hyperbola road model are estimated using multiple vanishing points located in image strips. The road tracking method is robust in dealing with complex road shapes, background clutters, shadows, and road markings. Experiments using real videos demonstrate the robustness of our approach.

Many approaches to the implementation of biometrics-based identification systems are possible, and different configurations are likely to generate significantly different operational characteristics. The choice of implementational structure is therefore very dependent on the performance criteria which are most important in any particular task scenario. In this paper we evaluate the merits of using multimodal structures, and we investigate how fundamentally different strategies for implementation can increase the degree of choice available in achieving particular performance criteria. In particular, we illustrate the merits of an implementation based on a multiagent computational architecture as a means of achieving high performance levels when recognition accuracy is a principal criterion. We also set out the relative merits of this strategy in comparison with other commonly adopted approaches to practical system realization. In particular we propose and evaluate a novel approach to implementation of a multimodal system based on negotiating agents.

Computer-based assessment systems analysing the drawing responses from a test subject have been widely explored within the area of neuropsychological dysfunction diagnosis and rehabilitation monitoring. This study reports on the development of a quantitative marking system for the automated assessment of visuo-spatial neglect. Using a clinically established pencil-and-paper based method as a marking benchmark, a set of features are extracted and selected from a battery of computer-captured drawing tasks. Through the application of linear regression and data transformation, the novel system is shown to be effective in correlating against a clinically recognised scale, while simultaneously improving the efficiency of the testing process.

Fingerprint matching has been approached using various criteria based on different extracted features. However, robust and accurate fingerprint matching is still a challenging problem. In this paper, we propose an improved integrated method which operates by first suggesting a consensus matching function, which combines different matching criteria based on heterogeneous features. We then devise a genetically guided approach to optimise the consensus matching function for simultaneous fingerprint alignment and verification. Since different features usually offer complementary information about the matching task, the consensus function is expected to improve the reliability of fingerprint matching. A related motivation for proposing such a function is to build a robust criterion that can perform well over a variety of different fingerprint matching instances. Additionally, by employing the global search functionality of a genetic algorithm along with a local matching operation for population initialisation, we aim to identify the optimal or near optimal global alignment between two fingerprints. The proposed algorithm is evaluated by means of a series of experiments conducted on public domain collections of fingerprint images and compared with previous work. Experimental results show that the consensus function can lead to a substantial improvement in performance while the local matching operation helps to identify promising initial alignment configurations, thereby speeding up the verification process. The resulting algorithm is more accurate than several other proposed methods which have been implemented for comparison.

In this paper, we present a robust road detection and tracking method based on a condensation particle filter for real-time video-based navigation applications. The image is divided into horizontal strips, and vanishing point (VP) detection is performed on each image strip. We propose a method for estimating the density of road boundary line segments in the image so that VP detection in an image strip takes into account the detection results in the neighboring image strips. This use of contextual information for VP detection leads to more accurate detection results. The estimated road parameters are then used to initialize the condensation tracker. Experiments using real road videos demonstrate the robustness of our method to difficult road conditions due to the presence of partial occlusion, shadows, and road signs.

Clustering is inherently a difficult task, and is made even more difficult when the selection of relevant features is also an issue. In this paper we propose an approach for simultaneous clustering and feature selection using a niching memetic algorithm. Our approach (which we call NMA_CFS) makes feature selection an integral part of the global clustering search procedure and attempts to overcome the problem of identifying less promising locally optimal solutions in both clustering and feature selection, without making any a priori assumption about the number of clusters. Within the NMA_CFS procedure, a variable composite representation is devised to encode both feature selection and cluster centers with different numbers of clusters. Further, local search operations are introduced to refine feature selection and cluster centers encoded in the chromosomes. Finally, a niching method is integrated to preserve the population diversity and prevent premature convergence. In an experimental evaluation we demonstrate the effectiveness of the proposed approach and compare it with other related approaches, using both synthetic and real data.

The collection of human biometric test data for system development and evaluation within any chosen modality generally requires significant time and effort if data are to be obtained in workable quantities. To overcome this problem, techniques to generate synthetic data have been developed. This paper describes a novel technique for the automatic synthesis of human handwritten-signature images, which introduces modeled variability within the generated output based on positional variation that is naturally found within genuine source data. The synthesized data were found to generate similar verification rates to those obtained using genuine data with the use of a commercial verification engine, thereby indicating the suitability of the data synthesized by using this method for a wide range of application scenarios.

Biometric authentication is increasingly gaining popularity in a wide range of applications. However, the storage of the biometric templates and/or encryption keys that are necessary for such applications is a matter of serious concern, as the compromise of templates or keys necessarily compromises the information secured by those keys. In this paper, we propose a novel method, which requires storage of neither biometric templates nor encryption keys, by directly generating the keys from statistical features of biometric data. An outline of the process is as follows: given biometric samples, a set of statistical features is first extracted from each sample. On each feature subset or single feature, we model the intra and interuser variation by clustering the data into natural clusters using a fuzzy genetic clustering algorithm. Based on the modelling results, we subsequently quantify the consistency of each feature subset or single feature for each user. By selecting the most consistent feature subsets and/or single features for each user individually, we generate the key reliably without compromising its relative security. The proposed method is evaluated on handwritten signature data and compared with related methods, and the results are very promising.

We present a robust road detection and tracking method using multiple vanishing points and the condensation filter. We represent the road using an extended hyperbola model with an added non-linear term to handle transitions between straight and curved road segments. The parameters of the road model are estimated using multiple vanishing points located in road segments. A vanishing line is then determined using a robust iterative curve fitting technique to recover parameters of the road model. These are then fed into a robust condensation tracker [M. Isard, A. Blake, Condensation: conditional density propagation for visual tracking, Int. J. Comput. Vision, 1998] to track the road. The tracker is able to deal with difficult road conditions. Experiments using real road videos demonstrate the suitability of our approach for real-time applications. A comparison with the Kalman filtering technique demonstrates the robustness of our approach.