Dissertations / Theses on the topic 'Biometics'

Personal identity refers to a set of attributes (e.g., name, social insurance number, etc.) that are associated with a person. Identity management is the process of creating, maintaining and destroying identities of individuals in a population. Biometric technologies are technologies developed to use statistical analysis of an individual’s biological or behavioral traits to determine his identity. Biometrics based authentication systems offer a reliable solution for identity management, because of their uniqueness, relative stability over time and security (among other reasons). Public acceptance of biometric systems will depend on their ability to ensure robustness, accuracy and security. Although robustness and accuracy of such systems are rapidly improving, there still remain some issues of security and balancing it with privacy. While the uniqueness of biometric traits offers a convenient and reliable means of identification, it also poses the risk of unauthorized cross-referencing among databases using the same biometric trait. There is also a high risk in case of a biometric database being compromised, since it’s not possible to revoke the biometric trait and re-issue a new one as is the case with passwords and smart keys. This unique attribute of biometric based authentication system poses a challenge that might slow down public acceptance and the use of biometrics for authentication purposes in large scale applications. In this research we investigate the vulnerabilities of biometric systems focusing on template security in iris-based biometric recognition systems. The iris has been well studied for authentication purposes and has been proven accurate in large scale applications in several airports and border crossings around the world. The most widely accepted iris recognition systems are based on Daugman’s model that creates a binary iris template. In this research we develop different systems using watermarking, bio-cryptography as well as feature transformation to achieve revocability and security of binary templates in iris based biometric authentication systems, while maintaining the performance that enables widespread application of these systems. All algorithms developed in this research are applicable on already existing biometric authentication systems and do not require redesign of these existing, well established iris-based authentication systems that use binary templates.

The terrorist attacks of September 11, 2001 propelled the issue of aviation security to the forefront of the U.S. domestic agenda. Although hundreds of individual airports exist in the U.S., the travel activities at each of these airports combine to holistically comprise an aviation system that represents a significant portion of the U.S. social and economic infrastructure. Disruption at one airport resulting from a criminal act, such as terrorism, could exert detrimental effects upon the aviation system and U.S national security (9/11 Commission, 2004). Each U.S. airport is individually responsible for various aspects of security including the control of physical access to sensitive and secure areas and facilities (9/11 Commission, 2004). Biometric technology has been examined as one method of enhancing airport access control to mitigate the possibility of criminal acts against airports. However, successful implementation of biometric technology depends largely on how individual security directors at each airport perceive, understand, and accept that technology. Backgrounds, attitudes, and personal characteristics influence individual decisions about technology implementation (Rogers, 1995; Tornatzky and Fleischer, 1990). This study examines the problem of airport access control, as well as, the current trends in biometric technology. Utilizing a survey of airport security directors and security managers, this study draws upon innovation diffusion theory and organizational theories to determine what personal, organizational, and technical variables contribute to the propensity of airport security directors and managers to adopt biometric technology for airport access control.
Ph.D.
Department of Criminal Justice and Legal Studies
Health and Public Affairs
Public Affairs PhD

The knowledge of the leaf area is required every time the sugarcane is subjected to evaluation in which different varieties are compared or submitted to different environmental conditions, managements or fertilizations, therefore evidently showing alteration in dimensions of the plant leaves, so affecting the photosynthetic process. Although the higher breathing-photosynthesis balance increases the amount of the assimilates available to vegetative growth (Net assimilation rate), the preferential carbon allocation for the different fractions of the leaf-plant, stalks, and root system, that is conditioned by the medium and/or management, can change the growth indexes. Thus, the estimates of those indexes can supply subsidies for understanding the adaptations of the plants under different medium conditions: light, temperature, moisture and soil fertility. The growth analysis is considered as standard method, since it is fundamentally based on the fact that about 95% on average of the dry matter accumulated by plants during growth rather result from the photosynthetic activity (Benincasa, 2003). The remaining comes from the absorption of mineral nutrients. Although at lower amounts, the mineral nutrients are indispensable to the vegetal growth and development, since they are absorbed and incorporated into every vegetative biomass. On direct or indirect way, they participate into production of the organic compounds as well as in sucrose´, that is produced from the phosphates-trioses transported from the chloroplast to cytosol and is mainly exported to the vacuoles. The present study was carried out to evaluate the methods for estimation of the leaf area, population density, growth and yield of two sugarcane cultivars provided with high productive potential and fertilized with different phosphorus doses in the cane plant cycle. The randomized block experimental design was used with five replicates under a factorial scheme 2 x 6, and the treatments were composed by the cultivars RB867515 and RB92579 under six P doses (0, 30, 60, 90, 120 and 150 kg ha-1). At planting time, all plots were given amounts corresponding to 33 and 200 kg ha-1 N and K, respectively. When beginning the rainy season at 210 DAP, either side-dressing fertilization and the micronutrient fertilization at the doses 67; 6; 6; and 7 kg ha-1 N, Cu, Mn and Zn, respectively were accomplished along the cropping line. Nine evaluations were performed with plants collected within 2.4 m2 at 30-days average intervals for estimation of the leaf area by two methods, plant population and dry matter accumulation in either photosynthetically active leaves and the whole plant. The tenth sampling was accomplished at 458 DAP in order to determine the yield of the apparent sucrose. After obtaining the data, the statistical analysis was accomplished, by using the program SISVAR. The estimate of the leaf area by using just the leaf +3 is an alternative to the conventional estimate method, when determining the foliar area for sugarcane crop. The increased phosphate fertilization contributes to increment the leaf area in the sugarcane crop, specially for the RB92579 cv. The phosphate fertilization widely affected the leaf area index (IAF), dry matter accumulation in plant, crop growth rate (TCC), relative growth rate (TCR), net assimilation rate (TAL), and both stalk and sucrose yields. During the period from 280 to 310 DAP, the highest IAF and TCC occurred, whereas the RB92579 cv. showed higher values. The RB92579 cv., showed higher yield of dry matter, industrializable stalks and apparent sucrose.
Coordenação de Aperfeiçoamento de Pessoal de Nível Superior
O conhecimento da área foliar se faz necessário toda vez em que a cana-deaçúcar é submetida a avaliações, onde se comparam variedades diferentes, ou submete-as a condições ambientais, manejo ou de adubação diferentes, que possam manifestar notadamente alterações nas dimensões das folhas das plantas influindo no processo fotossintético. Maior balanço fotossínteserespiração, aumenta a quantidade de assimilados disponíveis para o crescimento vegetal (Taxa de assimilação líquida), entretanto, a alocação preferencial de carbono pelas diferentes frações da planta-folha, colmos e sistema radicular, condicionada pelo meio e/ou manejo, pode alterar os índices de crescimento. Desta forma, as estimativas desses índices podem fornecer subsídios para o entendimento das adaptações experimentadas pelas plantas sob diferentes condições de meio: luz, temperatura, umidade e fertilidade do solo. A análise do crescimento é considerada um método padrão pois se baseia fundamentalmente no fato de que cerca de 95%, em média, da matéria seca acumulada pelas plantas, ao longo do seu crescimento, resultam da atividade fotossintética (Benincasa, 2003). O restante provém da absorção de nutrientes minerais. Embora em quantidades de menor expressão, os nutrientes minerais são indispensável ao crescimento desenvolvimento vegetal, visto que os mesmos são absorvidos e incorporados em toda biomassa vegetal. Estes participam direta ou indiretamente na produção de compostos orgânicos, bem como de sacarose, que é produzida a partir de trioses-fosfatos transportadas do cloroplasto para o citossol é exportada principalmente para os vacúolos. O presente trabalho teve como objetivo estudar métodos de estimativa de área foliar, densidade populacional, crescimento e a produção de duas variedades de cana-de-açúcar de alto potencial produtivo adubadas com diferentes doses de fósforo, em ciclo de cana planta. O delineamento foi em blocos casualizados com cinco repetições em esquema fatorial 2 x 6, sendo os tratamentos compostos pelas variedades RB867515 e RB92579 sob seis doses de P (0, 30, 60, 90, 120 e 150 kg ha-1). Por ocasião do plantio, todas as parcelas receberam quantidades equivalentes a 33 e 200 kg ha-1 de N e K, respectivamente. No início do período chuvoso, aos 210 DAP, realizou-se a adubação de cobertura e adubação com micronutrientes nas doses de 67; 6; 6; e 7 kg ha-1 de N, Cu, Mn e Zn, respectivamente, ao longo da linha da cultura. Foram realizadas nove avaliações com coletas de plantas em 2,4 m2, em intervalos médios de 30 dias, para estimativas da área foliar, por dois métodos, da população de plantas e do acúmulo de matéria seca nas ativas folhas fotossinteticamente e em toda planta. A décima amostragem foi realizada aos 458 DAP para determinar a produção de sacarose aparente. De posse dos dados realizou-se a análise estatística por meio do programa SISVAR. A estimativa de área foliar apenas com utilização da folha +3, constitui-se em uma alternativa ao método convencional de estimativa, na determinação da área foliar para a cultura da cana-de-açúcar. O aumento da adubação fosfatada contribui com o incremento de área foliar na cultura da cana-de-açúcar, sobretudo para a variedade RB92579. A adubação fosfatada influenciou grandemente no índice de área foliar (IAF), acúmulo de matéria seca na planta, taxa de crescimento da cultura (TCC), taxa de crescimento relativo (TCR), taxa de assimilação líquida (TAL), e produção de colmos e de sacarose. Os maiores IAF e TCC, ocorreram no período de 280 a 310 DAP, tendo a variedade RB92579 apresentada valores maiores. A variedade RB92579, apresentou maiores produções de matéria seca, colmos industrializáveis e sacarose aparente.

The aim of this thesis is to provide comprehensive information on the possibilities of authentication, combination of authentication factors and the integration of this issue into corporate communication. The work focuses on this issue and specifies the possibilities for obtaining authentication information, analyses the authentication methods, identification and authorization. It examines the applicability of biometric technologies, the principle of their functionality, examples of their use, their impact, the advantages and disadvantages they bring. A natural, easy-to-use, convenient tool for effective and secure communication is authentication including the dynamic biometric signature. The issues of the dynamic biometric signature technology and its implementation are examined from a comprehensive perspective involving experiments. The research proved that the dynamic biometric signature can serve as a method for supporting secure corporate communication and reduce authentication risks in companies and for individuals.

Gait and face biometrics have a unique advantage in that they can be used when images are acquired at a distance and signals are at too low a resolution to be perceived by other biometrics. Given such situations, some traits can be difficult to extract automatically but can still be perceived semantically using human vision. It is contended that such semantic annotations are usable as soft biometric signatures, useful for identification tasks. Feature subset selection techniques are employed to compare the distinguishing ability of individual semantically described physical traits. Their identification ability is also explored, both in isolation and in the improvement of the recognition rates of some associated gait biometric signatures using fusion techniques. This is the first approach to explore semantic descriptions of physiological human traits as used alone or to complement primary biometric techniques to facilitate recognition and analysis of surveillance video. Potential traits to be described are explored and justified against their psychological and practical merits. A novel dataset of semantic annotations is gathered describing subjects in two existing biometric datasets. Two applications of these semantic features and their associated biometric signatures are explored using the data gathered. We also draw on our experiments as a whole to highlight those traits thought to be most useful in assisting biometric recognition overall. Effective analysis of surveillance data by humans relies on semantic retrieval of the data which has been enriched by semantic annotations. A manual annotation process is time-consuming and prone to error due to various factors. We explore the semantic content-based retrieval of surveillance captured subjects. Working under the premise that similarity of the chosen biometric signature implies similarity of certain semantic traits, a set of semantic retrieval experiments are performed using well established Latent Semantic Analysis techniques.

Identity verification systems are an important part of our every day life. A typical example is the Automatic Teller Machine (ATM) which employs a simple identity verification scheme: the user is asked to enter their secret password after inserting their ATM card; if the password matches the one prescribed to the card, the user is allowed access to their bank account. This scheme suffers from a major drawback: only the validity of the combination of a certain possession (the ATM card) and certain knowledge (the password) is verified. The ATM card can be lost or stolen, and the password can be compromised. Thus new verification methods have emerged, where the password has either been replaced by, or used in addition to, biometrics such as the person’s speech, face image or fingerprints. Apart from the ATM example described above, biometrics can be applied to other areas, such as telephone & internet based banking, airline reservations & check-in, as well as forensic work and law enforcement applications. Biometric systems based on face images and/or speech signals have been shown to be quite effective. However, their performance easily degrades in the presence of a mismatch between training and testing conditions. For speech based systems this is usually in the form of channel distortion and/or ambient noise; for face based systems it can be in the form of a change in the illumination direction. A system which uses more than one biometric at the same time is known as a multi-modal verification system; it is often comprised of several modality experts and a decision stage. Since a multi-modal system uses complimentary discriminative information, lower error rates can be achieved; moreover, such a system can also be more robust, since the contribution of the modality affected by environmental conditions can be decreased. This thesis makes several contributions aimed at increasing the robustness of single- and multi-modal verification systems. Some of the major contributions are listed below. The robustness of a speech based system to ambient noise is increased by using Maximum Auto-Correlation Value (MACV) features, which utilize information from the source part of the speech signal. A new facial feature extraction technique is proposed (termed DCT-mod2), which utilizes polynomial coefficients derived from 2D Discrete Cosine Transform (DCT) coefficients of spatially neighbouring blocks. The DCT-mod2 features are shown to be robust to an illumination direction change as well as being over 80 times quicker to compute than 2D Gabor wavelet derived features. The fragility of Principal Component Analysis (PCA) derived features to an illumination direction change is solved by introducing a pre-processing step utilizing the DCT-mod2 feature extraction. We show that the enhanced PCA technique retains all the positive aspects of traditional PCA (that is, robustness to compression artefacts and white Gaussian noise) while also being robust to the illumination direction change. Several new methods, for use in fusion of speech and face information under noisy conditions, are proposed; these include a weight adjustment procedure, which explicitly measures the quality of the speech signal, and a decision stage comprised of a structurally noise resistant piece-wise linear classifier, which attempts to minimize the effects of noisy conditions via structural constraints on the decision boundary.

Nowadays biometric identification is used in a variety of applications-administration, business and even home. Although there are a lot of biometric identifiers, fingerprints are the most widely spread due to their acceptance from the people and the cheap price of the hardware equipment. Fingerprint recognition is a complex image recognition problem and includes algorithms and procedures for image enhancement and binarization, extracting and matching features and sometimes classification. In this work the main approaches in the research area are discussed, demonstrated and tested in a sample application. The demonstration software application is developed by using Verifinger SDK and Microsoft Visual Studio platform. The fingerprint sensor for testing the application is AuthenTec AES2501.

Made available in DSpace on 2015-03-03T15:48:40Z (GMT). No. of bitstreams: 1 FernandoP_TESE.pdf: 2701691 bytes, checksum: 2a3af30ede2c717ab23b1c7dc03a128a (MD5) Previous issue date: 2013-05-24
This work discusses the application of techniques of ensembles in multimodal recognition systems development in revocable biometrics. Biometric systems are the future identification techniques and user access control and a proof of this is the constant increases of such systems in current society. However, there is still much advancement to be developed, mainly with regard to the accuracy, security and processing time of such systems. In the search for developing more efficient techniques, the multimodal systems and the use of revocable biometrics are promising, and can model many of the problems involved in traditional biometric recognition. A multimodal system is characterized by combining different techniques of biometric security and overcome many limitations, how: failures in the extraction or processing the dataset. Among the various possibilities to develop a multimodal system, the use of ensembles is a subject quite promising, motivated by performance and flexibility that they are demonstrating over the years, in its many applications. Givin emphasis in relation to safety, one of the biggest problems found is that the biometrics is permanently related with the user and the fact of cannot be changed if compromised. However, this problem has been solved by techniques known as revocable biometrics, which consists of applying a transformation on the biometric data in order to protect the unique characteristics, making its cancellation and replacement. In order to contribute to this important subject, this work compares the performance of individual classifiers methods, as well as the set of classifiers, in the context of the original data and the biometric space transformed by different functions. Another factor to be highlighted is the use of Genetic Algorithms (GA) in different parts of the systems, seeking to further maximize their eficiency. One of the motivations of this development is to evaluate the gain that maximized ensembles systems by different GA can bring to the data in the transformed space. Another relevant factor is to generate revocable systems even more eficient by combining two or more functions of transformations, demonstrating that is possible to extract information of a similar standard through applying different transformation functions. With all this, it is clear the importance of revocable biometrics, ensembles and GA in the development of more eficient biometric systems, something that is increasingly important in the present day
O presente trabalho aborda a aplica??o de t?cnicas de comit?s de classificadores no desenvolvimento de sistemas de reconhecimento multimodais em biometrias revog?veis. Sistemas biom?tricos s?o o futuro das t?cnicas de identifica??o e controle de acesso de usu?rios, prova disso, s?o os aumentos constantes de tais sistemas na sociedade atual. Por?m, ainda existem muitos avan?os a serem desenvolvidos, principalmente no que se refere ? acur?cia, seguran?a e tempo de processamento de tais sistemas. Na busca por desenvolver t?cnicas mais eficientes, os sistemas multimodais e a utiliza??o de biometrias revog?veis mostram-se promissores, podendo contornar muitos dos problemas envolvidos no reconhecimento biom?trico tradicional. Um sistema multimodal ? caracterizado por combinar diferentes t?cnicas de seguran?a biom?trica e com isso, superar muitas limita- ??es, como: falhas de extra??o ou processamento dos dados. Dentre as v?rias possibilidades de se desenvolver um sistema multimodal, a utiliza??o de comit?s de classificadores ? um assunto bastante promissor, motivado pelo desempenho e flexibilidade que os mesmos v?m demonstrando ao longo dos anos, em suas in?meras aplica??es. Dando ?nfase em rela- ??o ? seguran?a, um dos maiores problemas encontrados se deve as biometrias estarem relacionadas permanentemente com o usu?rio e o fato de n?o poderem ser alteradas caso comprometidas. No entanto, esse problema vem sendo solucionado por t?cnicas conhecidas como biometrias revog?veis, as quais consistem em aplicar uma transforma??o sobre os dados biom?tricos de forma a proteger as caracter?sticas originais, possibilitando seu cancelamento e substitui??o. Com o objetivo de contribuir com esse importante tema, esse trabalho compara o desempenho de m?todos de classifica??es individuais, bem como conjunto de classificadores, no contexto dos dados originais e no espa?o biom?trico transformado por diferentes fun??es. Outro fator a se destacar, ? o uso de Algoritmos Gen?ticos (AGs) em diferentes partes dos sistemas, buscando maximizar ainda mais a efici?ncia dos mesmos. Uma das motiva??es desse desenvolvimento ? avaliar o ganho que os sistemas de comit?s maximizados por diferentes AGs podem trazer aos dados no espa?o transformado. Tamb?m busca-se gerar sistemas revog?veis ainda mais eficientes, atrav?s da combina??o de duas ou mais fun??es de transforma??o revog?veis, demonstrando que ? poss?vel extrair informa??es complementares de um mesmo padr?o atrav?s de tais procedimentos. Com tudo isso, fica claro a import?ncia das biometrias revog?veis, comit?s de classificadores e AGs, no desenvolvimento de sistemas biom?tricos mais eficientes, algo que se mostra cada vez mais importante nos dias atuais

This diploma thesis focuses on multibiometric systems, specifically on biometric fusion. The thesis describes eye biometrics, i.e. recognition based on retina and iris. The key part consists of design and implementation specification of a biometric system based on retina and iris recognition.

Automatic feature extraction still remains a relevant image and signal processing problem even tough both the field and technologies are developing rapidly. Images of low quality, where it is extremely difficult to reliably process image information automatically, are of special interest. To such images we can refer forensic fingerprints, which are left unintentionally on different surfaces andare contaminated by several of the most difficult noise types. For this reason, identification of fingerprints is mainly based on the visual skills of forensic examiners. We address the problem caused by low quality in fingerprints by connecting different sources of information together, yielding dense frequency and orientation maps in an iterative scheme. This scheme comprises smoothing ofthe original, but only along, ideally never across, the ridges. Reliable estimation of dense maps allows to introduce a continuous fingerprint ridge counting technique. In fingerprint scenario the collection of irrefutable tiny details, e.g. bifurcation of ridges, called minutiae, is used to tie the pattern of such points and their tangential directions to the finger producing the pattern. This limited feature set, location and direction of minutiae, is used in current AFIS systems, while fingerprint examiners use the extended set of features, including the image information between the points. With reasonably accurate estimationsof dense frequency and orientation maps at hand, we have been able to propose a novel compact feature descriptor of arbitrary points. We have used these descriptors to show that the image information between minutiae can be extracted automatically and be valuable for identity establishment of forensic images even if the underlying images are noisy. We collect and compress the image information in the neighborhoods of the fine details, such as minutiae, to vectors, one per minutia, and use the vectors to "color" the minutiae. When matching two patterns (of minutiae) even the color of the minutia must match to conclude that they come from the same identity. This feature development has been concentrated and tested on forensic fingerprint images. However, we have also studied an extension of its application area to other biometrics, periocular regions of faces. This allowed us to test the persistence of automatically extracted features across different types of imagesand image qualities, supporting its generalizability.

Information security has always been a topic of concern in the world as an emphasis on new techniques to secure the identity of a legitimate user is regarded as top priority. To counter such an issue, we have a traditional way of authentication factors “what you have” and “what you know” in the form of smart cards or passwords respectively. But biometrics is based on the factor “who are you” by analyzing human physical or behavioral characteristics. Biometrics has always been an efficient way of authorization and is now considered as a $1500 million industry where fingerprints dominate the biometrics while iris is quickly emerging as the most desirable form of biometric technique.The main goal of this thesis is to compare and evaluate different biometrics techniques in terms of their purpose, recognition mechanism, market value and their application areas. Since there are no defined evaluating criteria, my method of evaluation was based on a literature survey from internet, books, IEEE papers and technical surveys. Chapter 3 is focused on different biometrics techniques where I discuss them briefly but in chapter 4, I go deeper into Iris, fingerprints, facial techniques which are prominent in biometrics world. Lastly, I had a general assessment of the biometrics, their future growth and suggested specific techniques for different environment like access controls, e-commerce, national ids, and surveillance.

Gait biometrics has attracted increasing interest in the computer vision and machine learning communities because of its unique advantages for recognition at distance. However, there have as yet been few gait biometric approaches which use temporal three-dimensional(3D) data. Clearly, 3D gait data conveys more information than 2D gait data and it is also the natural representation of human gait as perceived by humans.The University of Southampton has created a multi-biometric tunnel using twelve cameras to capture multiple gait images and reconstruct them into 3D volumetric gait data. Some analyses have been done using this 3D dataset mainly to solve the view dependent problem using model-free silhouette-based approaches. This thesis explores the potential of model-based methods in an indoor 3D volumetric gait dataset and presents a novel human gait features extraction algorithm based on marionette and mass-spring principles. We have developed two different model-based approaches to extract human gait kinematics from 3D volumetric gait data. The first approach used a structural model of a human. This model contained four articulated cylinders and four joints with two degrees of rotational freedom at each joint to model the human lower legs. Human gait kinematic trajectories were extracted by fitting the gait model to the gait data. We proposed a simple yet effective model-fitting algorithm using a correlation filter and dynamic programming. To increase the fitting performance, we utilized a genetic algorithm on top of this structural model. The second approach was a novel 3D model-based approach using a marionette-based mass-spring model. To model the articulated human body, we used a stick-figure model which emulates marionette's motion and joint structure. The stick-figure model had eleven nodes representing the human joints of head, torso and lower legs. Each node was linked with at least one other node by spring. The voxel data in the next frame had a role as an attractor which able to generate forces for each node and then iteratively warp the model into the data. This process was repeated for successive frames. Our methods can extract both structural and dynamic gait features. Some of the extracted features were inherently unique to 3D gait data such as footprint angle and pelvis rotation. Analysis on a database of 46 subjects shows an encouraging correct classification rate up to 95.1% and suggests that model-based 3D gait analysis can contribute even more in gait biometrics.

Humans naturally use descriptions to verbally convey the appearance of an individual. Eyewitness descriptions are an important resource for many criminal investigations. However, they cannot be used to automatically search databases featuring video or biometric data - reducing the utility of human descriptions in the search for the suspect. Soft biometrics are a new form of biometric identification which uses physical or behavioural traits that can be naturally described by humans. This thesis will explore how soft biometrics can be used alongside traditional biometrics, allowing video footage and biometric data to be searched using a description. To permit soft biometric identification the human description must be accurate, yet conventional descriptions comprising of absolute labels and estimations are often unreliable. A novel method of obtaining human descriptions will be introduced which utilizes comparative categorical labels to describe the differences between subjects. A database of facial and bodily comparative labels is introduced and analysed. Prior to use as a biometric feature, comparative descriptions must be anchored. Several techniques to convert multiple comparative labels into a single relative measurement are explored. Recognition experiments were conducted to assess the discriminative capabilities of relative measurements as a biometric. Relative measurements can also be obtained from other forms of human representation. This is demonstrated using several machine learning techniques to determine relative measurements from gait biometric signatures. Retrieval results are presented showing the ability to automatically search video footage using comparative descriptions.

Aujourd'hui, il existe de multiples usages des systèmes biométriques à de nombreuses fins telles que le contrôle d'accès physique, le contrôle de présence, le paiement électronique et autres. Cette thèse de doctorat porte sur l'authentification biométrique et nous proposons d'utiliser la dynamique de frappe au clavier afin d'éviter les problèmes d'authentification par mot de passe. La dynamique de frappe au clavier mesure les rythmes qui se dégagent lorsqu'on tape sur un clavier d'ordinateur. En ce sens, c'est une modalité biométrique comportementale, de même que la dynamique de signature, la démarche ou la voix. Parmi les avantages de la dynamique de frappe au clavier par rapport à d'autres modalités, nous pouvons mentionner son faible coût et sa facilité d'usage : en effet, aucun capteur ni dispositif supplémentaire n'est nécessaire et les utilisateurs sont habitués à taper un mot de passe. En contrepartie, la dynamique de frappe présente de plus faibles performances que les autres modalités biométriques comme les empreintes digitales, le visage, l'iris. Cela peut s'expliquer par une variabilité intra-classe élevée. Une façon de gérer cette variabilité est de prendre en compte des informations supplémentaires dans le processus de décision. Cela peut être fait de différentes manières : (i) en combinant la dynamique de frappe au clavier avec une autre modalité biométrique (multibiométrie); (ii) en optimisant l'étape d'enrôlement (une donnée biométrique est exploitée pour la génération de la référence seulement si le niveau de qualité est suffisant); ou (iii) avec une solution nouvelle et prometteuse: la biométrie douce (profilage de l'utilisateur). Nous abordons dans cette thèse ces deux derniers aspects. Nous proposons plusieurs contributions afin d'améliorer les performances des systèmes de dynamique de frappe au clavier. Tout d'abord, nous avons créé notre propre jeu de données, qui est une nouvelle base de données biométrique appelée 'GREYC-NISLAB Keystroke'. Nous avons collecté les données de 110 utilisateurs en France et en Norvège. Cette nouvelle base est publique et contient des informations de profilage des utilisateurs: la façon de taper (une main ou deux mains), le genre, l'âge et la latéralité manuelle (droiter ou gaucher). Nous avons effectué diverses études afin de déterminer le taux de reconnaissance des critères de biométrie douce : (i) la façon de taper (une main ou deux mains); (ii) le genre (masculin ou féminin); (iii) la classe d'âge (moins de 30 ans ou plus de 30 ans); et (iv) la latéralité manuelle (droitier ou gaucher) des utilisateurs en fonction de leur façon de taper au clavier. Nous montrons qu'il est possible de reconnaître le profil de l'utilisateur en fonction de ces critères. Par la suite, nous proposons une fusion de différentes acquisitions de la dynamique de frappe afin d'accroître les performances du système. Enfin, en combinant les processus d'authentification avec les profils de biométrie douce, nous présentons une amélioration de l'authentification. Les résultats de nos expériences montrent les avantages des méthodes proposées
At present, there are a number of usages of biometric systems for many specific purposes such as physical access control, attendance monitoring, electronic payment (e-payment) and others. This PhD thesis focuses on biometric authentication and we propose to use keystroke dynamics in order to avoid password-based authentication problems. Keystroke dynamics measures the rhythm a person exhibits while typing on a keyboard. In this sense, keystroke dynamics is a behavioral biometric modality, as well as signature dynamics, gait and voice. Among the advantages of keystroke dynamics in comparison to other modalities, we can mention that it is a low cost and usable modality: indeed, no extra sensor or device is required and users often type a password. The counterpart to these advantages is the worse performance compared to morphological biometric modalities such as fingerprint, face or iris. The rather worse performances of keystroke dynamics can be explained by the high intra-class variability of the users' behaviour. One way to handle this variability is to take into account additional information in the decision process. This can be done with: (i) multibiometrics (by combining keystroke and another modality); (ii) optimising the enrolment step (a template is stored as reference only if its quality level is sufficient); or (iii) with a new and promising solution: soft biometrics (profiling the user). We address in this PhD thesis these two last aspects. We propose several contributions in order to enhance the performance of keystroke dynamics systems. First, we created a benchmark dataset called 'GREYC-NISLAB Keystroke' with biometric data collection from 110 users in France and Norway. This new benchmark database is available to the international scientific community and contains some profiling information on users: the way of typing (one hand or two hands), gender, age and handedness. We then perform various studies in order to determine the recognition accuracy of soft biometric traits given keystroke dynamics features: (i) the way of typing (one hand or two hands); (ii) gender (male or female); (iii) age class (below 30 or 30 and above); and (iv) handedness (right-handed or left-handed). Subsequently, we study the biometric fusion with keystroke dynamics in order to increase the soft biometrics recognition performance. Finally, by combining the authentication process with soft criteria, we present an improvement of user verification. The results of our experiments show the benefits of the proposed methods

Biometric systems allow identification of human persons based on physiological or behavioral characteristics, such as voice, handprint, iris or facial characteristics. The use of face and iris recognition as a way to authenticate user’s identities has been a topic of research for years. Present iris recognition systems require that subjects stand close (<2m) to the imaging camera and look for a period of about three seconds until the data are captured. This cooperative behavior is required in order to capture quality images for accurate recognition. This will eventually restrict the amount of practical applications where iris recognition can be applied, especially in an uncontrolled environment where subjects are not expected to cooperate such as criminals and terrorists, for example. For this reason, this thesis develops a collection of methods to deal with low quality face and iris images and that can be applied for face and iris recognition in a non-cooperative environment. This thesis makes the following main contributions: I. For eye and face tracking in low quality images, a new robust method is developed. The proposed system consists of three parts: face localization, eye detection and eye tracking. This is accomplished using traditional image-based passive techniques such as shape information of the eye and active based methods which exploit the spectral properties of the pupil under IR illumination. The developed method is also tested on underexposed images where the subject shows large head movements. II. For iris recognition, a new technique is developed for accurate iris segmentation in low quality images where a major portion of the iris is occluded. Most existing methods perform generally quite well but tend to overestimate the occluded regions, and thus lose iris information that could be used for identification. This information loss is potentially important in the covert surveillance applications we consider in this thesis. Once the iris region is properly segmented using the developed method, the biometric feature information is calculated for the iris region using the relative entropy technique. Iris biometric feature information is calculated using two different feature decomposition algorithms based on Principal Component Analysis (PCA) and Independent Component Analysis (ICA). III. For face recognition, a new approach is developed to measure biometric feature information and the changes in biometric sample quality resulting from image degradations. A definition of biometric feature information is introduced and an algorithm to measure it proposed, based on a set of population and individual biometric features, as measured by a biometric algorithm under test. Examples of its application were shown for two different face recognition algorithms based on PCA (Eigenface) and Fisher Linear Discriminant (FLD) feature decompositions.

The research conducted under this study offers an understanding of the reasons why information technology (IT) and/or information assurance (IA) managers choose to recommend or not to recommend particular technologies, specifically biometric security, to their organizations. A review of the relevant literature provided the foundation to develop a set of research questions and factors for this research effort. The research questions became the basis of the study’s stated hypotheses for examining managers’ perceptions of the security effectiveness, need, reliability, and cost-effectiveness of biometrics. The research indicates that positive perceptions of security effectiveness, need, reliability, and cost-effectiveness correlate with IT/IA managers’ willingness to recommend biometric security technologies. The implications of this study are that executives and managers can make informed decisions about the recommendation and adoption process relevant to biometric security technologies through an understanding of how perceptions of biometric technology affect the decision to recommend this type of technology. The study’s results may also help biometric product developers, vendors, and marketers understand the important perceptions of biometric security technologies within their customer base of IT/IA managers.

This thesis describes research into the score-level fusion process in multimodal biometrics. The emphasis of the research is on the fusion of face and voice biometrics in the two recognition modes of verification and open-set identification. The growing interest in the use of multiple modalities in biometrics is due to its potential capabilities for eradicating certain important limitations of unimodal biometrics. One of the factors important to the accuracy of a multimodal biometric system is the choice of the technique deployed for data fusion. To address this issue, investigations are carried out into the relative performance of several statistical data fusion techniques for combining the score information in both unimodal and multimodal biometrics (i.e. speaker and/ or face verification). Another important issue associated with any multimodal technique is that of variations in the biometric data. Such variations are reflected in the corresponding biometric scores, and can thereby adversely influence the overall effectiveness of multimodal biometric recognition. To address this problem, different methods are proposed and investigated. The first approach is based on estimating the relative quality aspects of the test scores and then passing them on into the fusion process either as features or weights. The approach provides the possibility of tackling the data variations based on adjusting the weights for each of the modalities involved according to its relative quality. Another approach considered for tackling the effects of data variations is based on the use of score normalisation mechanisms. Whilst score normalisation has been widely used in voice biometrics, its effectiveness in other biometrics has not been previously investigated. This method is shown to considerably improve the accuracy of multimodal biometrics by appropriately correcting the scores from degraded modalities prior to the fusion process. The investigations in this work are also extended to the combination of score normalisation with relative quality estimation. The experimental results show that, such a combination is more effective than the use of only one of these techniques with the fusion process. The thesis presents a thorough description of the research undertaken, details the experimental results and provides a comprehensive analysis of them.

This thesis describes research into effective voice biometrics (speaker recognition) under mismatched noise conditions. Over the last two decades, this class of biometrics has been the subject of considerable research due to its various applications in such areas as telephone banking, remote access control and surveillance. One of the main challenges associated with the deployment of voice biometrics in practice is that of undesired variations in speech characteristics caused by environmental noise. Such variations can in turn lead to a mismatch between the corresponding test and reference material from the same speaker. This is found to adversely affect the performance of speaker recognition in terms of accuracy. To address the above problem, a novel approach is introduced and investigated. The proposed method is based on minimising the noise mismatch between reference speaker models and the given test utterance, and involves a new form of Test-Normalisation (T-Norm) for further enhancing matching scores under the aforementioned adverse operating conditions. Through experimental investigations, based on the two main classes of speaker recognition (i.e. verification/ open-set identification), it is shown that the proposed approach can significantly improve the performance accuracy under mismatched noise conditions. In order to further improve the recognition accuracy in severe mismatch conditions, an approach to enhancing the above stated method is proposed. This, which involves providing a closer adjustment of the reference speaker models to the noise condition in the test utterance, is shown to considerably increase the accuracy in extreme cases of noisy test data. Moreover, to tackle the computational burden associated with the use of the enhanced approach with open-set identification, an efficient algorithm for its realisation in this context is introduced and evaluated. The thesis presents a detailed description of the research undertaken, describes the experimental investigations and provides a thorough analysis of the outcomes.

The iris of the eye is considered to be the most discriminatory of facial features. It is often assumed that an individual's iris pattem remains unchanged throughout their lifetime. However, clinical findings suggest that changes in the iris can occur due to ageing and in response to external factors such as medications, disease and surgery. Such changes in the appearance of the iris need to be addressed when considering the iris as a biometric. The extent of change over time and whether this affects the appearance of the iris sufficiently to impact on its value as a biometric measure requires investigation. To enable a detailed study of iris stability, a novel database of high resolution iris images is presented comprising 364 irides with images captured at regular time intervals. To facilitate iris analysis, an enhanced localisation method is proposed for iris segmentation. An extensive evaluation of a number of feature extraction algorithms is described and applied to a database of iris images captured over increasing time intervals of three, six, nine and twelve months. Irides are analysed in order to determine whether significant variation exists between images captured at increasing time lapses and to assess the impact of such variations on recognition performance. Iris matching schemes are also evaluated and compared. Results are presented in which increased dissimilarity is observed in iris comparisons as the time interval between comparisons increases. These differences increase the probability of recognition failure in iris recognition. A greater number of recognition failures are observed as the time lapse between images increases. Such failures are also found to differ depending on iris texture pattern. It is concluded that whilst the iris remains an appropriate biometric identifier, it may not be as stable as originally proposed. Further research is required to determine the causes of the observed longitudinal variation in iris pattem and the possible impact on iris recognition systems.

One of the major challenges in human-machine interaction has always been the development of such techniques that are able to provide accurate human recognition, so as to other either personalized services or to protect critical infrastructures from unauthorized access. To this direction, a series of well stated and efficient methods have been proposed mainly based on biometric characteristics of the user. Despite the significant progress that has been achieved recently, there are still many open issues in the area, concerning not only the performance of the systems but also the intrusiveness of the collecting methods. The current thesis deals with the investigation of novel, activity-related biometric traits and their potential for multiple and unobtrusive authentication based on the spatiotemporal analysis of human activities. In particular, it starts with an extensive bibliography review regarding the most important works in the area of biometrics, exhibiting and justifying in parallel the transition that is performed from the classic biometrics to the new concept of behavioural biometrics. Based on previous works related to the human physiology and human motion and motivated by the intuitive assumption that different body types and different characters would produce distinguishable, and thus, valuable for biometric verification, activity-related traits, a new type of biometrics, the so-called prehension biometrics (i.e. the combined movement of reaching, grasping activities), is introduced and thoroughly studied herein. The analysis is performed via the so-called Activity hyper-Surfaces that form a dynamic movement-related manifold for the extraction of a series of behavioural features. Thereafter, the focus is laid on the extraction of continuous soft biometric features and their efficient combination with state-of-the-art biometric approaches towards increased authentication performance and enhanced security in template storage via Soft biometric Keys. In this context, a novel and generic probabilistic framework is proposed that produces an enhanced matching probability based on the modelling of the systematic error induced during the estimation of the aforementioned soft biometrics and the efficient clustering of the soft biometric feature space. Next, an extensive experimental evaluation of the proposed methodologies follows that effectively illustrates the increased authentication potential of the prehension-related biometrics and the significant advances in the recognition performance by the probabilistic framework. In particular, the prehension biometrics related biometrics is applied on several databases of ~100 different subjects in total performing a great variety of movements. The carried out experiments simulate both episodic and multiple authentication scenarios, while contextual parameters, (i.e. the ergonomic-based quality factors of the human body) are also taken into account. Furthermore, the probabilistic framework for augmenting biometric recognition via soft biometrics is applied on top of two state-of-art biometric systems, i.e. a gait recognition (> 100 subjects)- and a 3D face recognition-based one (~55 subjects), exhibiting significant advances to their performance. The thesis is concluded with an in-depth discussion summarizing the major achievements of the current work, as well as some possible drawbacks and other open issues of the proposed approaches that could be addressed in future works.

The growing presence of the Internet in day-to-day tasks, along with the evolution of computational systems, contributed to increase data exposure. This scenario highlights the need for safer user authentication systems. An alternative to deal with this is by the use of biometric systems. However, biometric features may change over time, an issue that can affect the recognition performance due to an outdated biometric reference. This effect can be called as template ageing in the area of biometrics and as concept drift in machine learning. It raises the need to automatically adapt the biometric reference over time, a task performed by adaptive biometric systems. This thesis studied adaptive biometric systems considering biometrics in a data stream context. In this context, the test is performed on a biometric data stream, in which the query samples are presented one after another to the biometric system. An adaptive biometric system then has to classify each query and adapt the biometric reference. The decision to perform the adaptation is taken by the biometric system. Among the biometric modalities, this thesis focused on behavioural biometrics, particularly on keystroke dynamics and on accelerometer biometrics. Behavioural modalities tend to be subject to faster changes over time than physical modalities. Nevertheless, there were few studies dealing with adaptive biometric systems for behavioural modalities, highlighting a gap to be explored. Throughout the thesis, several aspects to enhance the design of adaptive biometric systems for behavioural modalities in a data stream context were discussed: proposal of adaptation strategies for the immune-based classification algorithm Self-Detector, combination of genuine and impostor models in the Enhanced Template Update framework and application of score normalization to adaptive biometric systems. Based on the investigation of these aspects, it was observed that the best choice for each studied aspect of the adaptive biometric systems can be different depending on the dataset and, furthermore, depending on the users in the dataset. The different user characteristics, including the way that the biometric features change over time, suggests that adaptation strategies should be chosen per user. This motivated the proposal of a modular adaptive biometric system, named ModBioS, which can choose each of these aspects per user. ModBioS is capable of generalizing several baselines and proposals into a single modular framework, along with the possibility of assigning different adaptation strategies per user. Experimental results showed that the modular adaptive biometric system can outperform several baseline systems, while opening a number of new opportunities for future work.
A crescente presença da Internet nas tarefas do dia a dia, juntamente com a evolução dos sistemas computacionais, contribuiu para aumentar a exposição dos dados. Esse cenário evidencia a necessidade de sistemas de autenticação de usuários mais seguros. Uma alternativa para lidar com isso é pelo uso de sistemas biométricos. Contudo, características biométricas podem mudar com o tempo, o que pode afetar o desempenho de reconhecimento devido a uma referência biométrica desatualizada. Esse efeito pode ser chamado de template ageing na área de sistemas biométricos adaptativos ou de mudança de conceito em aprendizado de máquina. Isso levanta a necessidade de adaptar automaticamente a referência biométrica com o tempo, uma tarefa executada por sistemas biométricos adaptativos. Esta tese estudou sistemas biométricos adaptativos considerando biometria em um contexto de fluxo de dados. Neste contexto, o teste é executado em um fluxo de dados biométrico, em que as amostras de consulta são apresentadas uma após a outra para o sistema biométrico. Um sistema biométrico adaptativo deve então classificar cada consulta e adaptar a referência biométrica. A decisão de executar a adaptação é tomada pelo sistema biométrico. Dentre as modalidades biométricas, esta tese foca em biometria comportamental, em particular em dinâmica da digitação e em biometria por acelerômetro. Modalidades comportamentais tendem a ser sujeitas a mudanças mais rápidas do que modalidades físicas. Entretanto, havia poucos estudos lidando com sistemas biométricos adaptativos para modalidades comportamentais, destacando uma lacuna para ser explorada. Ao longo da tese, diversos aspectos para aprimorar o projeto de sistemas biométricos adaptativos para modalidades comportamentais em um contexto de fluxo de dados foram discutidos: proposta de estratégias de adaptação para o algoritmo de classificação imunológico Self-Detector, combinação de modelos genuíno e impostor no framework do Enhanced Template Update e aplicação de normalização de scores em sistemas biométricos adaptativos. Com base na investigação desses aspectos, foi observado que a melhor escolha para cada aspecto estudado dos sistemas biométricos adaptativos pode ser diferente dependendo do conjunto de dados e, além disso, dependendo dos usuários no conjunto de dados. As diferentes características dos usuários, incluindo a forma como as características biométricas mudam com o tempo, sugerem que as estratégias de adaptação deveriam ser escolhidas por usuário. Isso motivou a proposta de um sistema biométrico adaptativo modular, chamado ModBioS, que pode escolher cada um desses aspectos por usuário. O ModBioS é capaz de generalizar diversos sistemas baseline e propostas apresentadas nesta tese em um framework modular, juntamente com a possibilidade de atribuir estratégias de adaptação diferentes por usuário. Resultados experimentais mostraram que o sistema biométrico adaptativo modular pode superar diversos sistemas baseline, enquanto que abre um grande número de oportunidades para trabalhos futuros.

The aim of this semester project is to design a suitable database structure for the audio-visual data, parts extracted from the data corresponding to secondary biometric and non-biometric characteristics and description of these characters. The resulting database will be used for research into the identification of audio-visual content of multimedia data.

The aim of the thesis is to provide complete information regarding electronic documents and possibilities of their usage. The focus is concentrated on the area of authentication, which specifies the possibility of obtaining authentication information and describes the authentication processes itself. The diploma thesis also deals with the suggestion of multifactor authentication of electronic documents for the selected company.

Scientific Bird Ringing has been conducted for over 100 years. The data collected has been used for numerous purposes, studying population trends of particular species, discovering migratory routes, survival rates and assessing climate change and environmental pollution by using birds as bioindicators. Constant Effort ringing Sites are now used to collect years of data in a standardized way, making it possible to compare results with other sites and better study bird trends and communities. Portugal has hundreds of thousands of ringing records covering decades of data, most of these records are not being used or published. From the data collected in one Constant Effort Site in Herdade da Mitra (Portugal) we show how to sex different species based on biometric data, assess breeding site fidelity in migratory species, population trends for the study area and longevity records, as well as provide suggestions to improve ringing in the future; Resumo: A anilhagem científica de aves é feita há mais de 100 anos. Estes dados têm sido usados de várias formas, para estudar tendências populacionais de uma espécie, descobrir rotas migratórias, taxas de sobrevivência e para avaliar alterações climáticas e níveis de poluição através do uso de aves como bioindicadores. Estações de Esforço Constante recolhem dados durante anos, de forma padronizada, sendo possível comparar resultados com outras estações e estudar melhor as comunidades de aves. Portugal tem centenas de milhares de dados de anilhagem recolhidas durante décadas, a maioria destes dados não são usados ou publicados. A partir dos dados recolhidos na Estação da Herdade da Mitra (Portugal), mostramos como determinar o sexo de diferentes espécies de passeriformes com base em dados biométricos, avaliamos filopatria em aves migradoras, tendências populacionais para a área de estudo e recordes de longevidade, também são dadas sugestões para melhorar a anilhagem no futuro.

As online business has been very popular in the past decade, the tasks of providing user authentication and verification have become more important than before to protect user sensitive information from malicious hands. The most common approach to user authentication and verification is the use of password. However, the dilemma users facing in traditional passwords becomes more and more evident: users tend to choose easy-to-remember passwords, which are often weak passwords that are easy to crack. Meanwhile, behavioral biometrics have promising potentials in meeting both security and usability demands, since they authenticate users by "who you are", instead of "what you have". In this dissertation, we first develop two such user verification applications based on behavioral biometrics: the first one is via mouse movements, and the second via tapping behaviors on smartphones; then we focus on modeling user web browsing behaviors by Fitts' Law.;Specifically, we develop a user verification system by exploiting the uniqueness of people's mouse movements. The key feature of our system lies in using much more fine-grained (point-by-point) angle-based metrics of mouse movements for user verification. These new metrics are relatively unique from person to person and independent of the computing platform. We conduct a series of experiments to show that the proposed system can verify a user in an accurate and timely manner, and induced system overhead is minor. Similar to mouse movements, the tapping behaviors of smartphone users on touchscreen also vary from person to person. We propose a non-intrusive user verification mechanism to substantiate whether an authenticating user is the true owner of the smartphone or an impostor who happens to know the passcode. The effectiveness of the proposed approach is validated through real experiments. to further understand user pointing behaviors, we attempt to stress-test Fitts' law in the "wild", namely, under natural web browsing environments, instead of restricted laboratory settings in previous studies. Our analysis shows that, while the averaged pointing times follow Fitts' law very well, there is considerable deviations from Fitts' law. We observe that, in natural browsing, a fast movement has a different error model from the other two movements. Therefore, a complete profiling on user pointing performance should be done in more details, for example, constructing different error models for slow and fast movements. as future works, we plan to exploit multiple-finger tappings for smartphone user verification, and evaluate user privacy issues in Amazon wish list.

The overall objective in defining feature space is to reduce the dimensionality of the original pattern space, whilst maintaining discriminatory power for classification. To meet this objective in the context of ear biometrics a novel force field transformation is introduced in which the image is treated as an array of mutually attracting particles that act as the source of a Gaussian force field. In a similar way to Newton’s Law of Universal Gravitation pixels are imagined to attract each other according to the product of their intensities and inversely to the square of the distance between them. Underlying the force field there is a scalar potential energy field, which in the case of an ear takes the form of a smooth surface that resembles a small mountain with a number of peaks joined by ridges. The peaks correspond to potential energy wells and to extend the analogy the ridges correspond to potential energy channels. The directional properties of the force field are exploited to automatically locate these wells and channels, which then form the basis of a set of characteristic ear features. The new features are robust especially in the presence of noise, and have the advantage that the ear does not need to be explicitly extracted from its background. The directional properties of the ensuing force field lead to two equivalent extraction techniques; one is algorithmic and based on field lines, while the other is analytical and based on the divergence of force direction. The technique is validated by performing recognition on a database of ears selected from the XM2VTS face database. This confirms not only that ears do indeed appear to have potential as a biometric, but also that the new approach is well suited to their description.

L’augmentation rapide de la population combinée à la mobilité croissante des individus a engendré le besoin de systèmes de gestion d’identités sophistiqués. À cet effet, le terme biométrie se réfère généralement aux méthodes permettant d’identifier les individus en utilisant des caractéristiques biologiques ou comportementales. Les méthodes les plus populaires, c’est-à-dire la reconnaissance d’empreintes digitales, d’iris ou de visages, se basent toutes sur des méthodes de vision par ordinateur. L’adoption de réseaux convolutifs profonds, rendue possible par le calcul générique sur processeur graphique, ont porté les récentes avancées en vision par ordinateur. Ces avancées ont permis une amélioration drastique des performances des méthodes conventionnelles en biométrie, ce qui a accéléré leur adoption pour des usages concrets, et a provoqué un débat public sur l’utilisation de ces techniques. Dans ce contexte, les concepteurs de systèmes biométriques sont confrontés à un grand nombre de challenges dans l’apprentissage de ces réseaux. Dans cette thèse, nous considérons ces challenges du point de vue de l’apprentissage statistique théorique, ce qui nous amène à proposer ou esquisser des solutions concrètes. Premièrement, nous répondons à une prolifération de travaux sur l’apprentissage de similarité pour les réseaux profonds, qui optimisent des fonctions objectif détachées du but naturel d’ordonnancement recherché en biométrie. Précisément, nous introduisons la notion d’ordonnancement par similarité, en mettant en évidence la relation entre l’ordonnancement bipartite et la recherche d’une similarité adaptée à l’identification biométrique. Nous étendons ensuite la théorie sur l’ordonnancement bipartite à ce nouveau problème, tout en l’adaptant aux spécificités de l’apprentissage sur paires, notamment concernant son coût computationnel. Les fonctions objectif usuelles permettent d’optimiser la performance prédictive, mais de récents travaux ont mis en évidence la nécessité de prendre en compte d’autres facteurs lors de l’entraı̂nement d’un système biométrique, comme les biais présents dans les données, la robustesse des prédictions ou encore des questions d’équité. La thèse aborde ces trois exemples, en propose une étude statistique minutieuse, ainsi que des méthodes pratiques qui donnent les outils nécessaires aux concepteurs de systèmes biométriques pour adresser ces problématiques, sans compromettre la performance de leurs algorithmes
The rapid growth in population, combined with the increased mobility of people has created a need for sophisticated identity management systems.For this purpose, biometrics refers to the identification of individuals using behavioral or biological characteristics. The most popular approaches, i.e. fingerprint, iris or face recognition, are all based on computer vision methods. The adoption of deep convolutional networks, enabled by general purpose computing on graphics processing units, made the recent advances incomputer vision possible. These advances have led to drastic improvements for conventional biometric methods, which boosted their adoption in practical settings, and stirred up public debate about these technologies. In this respect, biometric systems providers face many challenges when learning those networks.In this thesis, we consider those challenges from the angle of statistical learning theory, which leads us to propose or sketch practical solutions. First, we answer to the proliferation of papers on similarity learningfor deep neural networks that optimize objective functions that are disconnected with the natural ranking aim sought out in biometrics. Precisely, we introduce the notion of similarity ranking, by highlighting the relationship between bipartite ranking and the requirements for similarities that are well suited to biometric identification. We then extend the theory of bipartite ranking to this new problem, by adapting it to the specificities of pairwise learning, particularly those regarding its computational cost. Usual objective functions optimize for predictive performance, but recentwork has underlined the necessity to consider other aspects when training a biometric system, such as dataset bias, prediction robustness or notions of fairness. The thesis tackles all of those three examplesby proposing their careful statistical analysis, as well as practical methods that provide the necessary tools to biometric systems manufacturers to address those issues, without jeopardizing the performance of their algorithms

It is commonly acknowledged that mobile devices now form an integral part of an individual’s everyday life. The modern mobile handheld devices are capable to provide a wide range of services and applications over multiple networks. With the increasing capability and accessibility, they introduce additional demands in term of security. This thesis explores the need for authentication on mobile devices and proposes a novel mechanism to improve the current techniques. The research begins with an intensive review of mobile technologies and the current security challenges that mobile devices experience to illustrate the imperative of authentication on mobile devices. The research then highlights the existing authentication mechanism and a wide range of weakness. To this end, biometric approaches are identified as an appropriate solution an opportunity for security to be maintained beyond point-of-entry. Indeed, by utilising behaviour biometric techniques, the authentication mechanism can be performed in a continuous and transparent fashion. This research investigated three behavioural biometric techniques based on SMS texting activities and messages, looking to apply these techniques as a multi-modal biometric authentication method for mobile devices. The results showed that linguistic profiling; keystroke dynamics and behaviour profiling can be used to discriminate users with overall Equal Error Rates (EER) 12.8%, 20.8% and 9.2% respectively. By using a combination of biometrics, the results showed clearly that the classification performance is better than using single biometric technique achieving EER 3.3%. Based on these findings, a novel architecture of multi-modal biometric authentication on mobile devices is proposed. The framework is able to provide a robust, continuous and transparent authentication in standalone and server-client modes regardless of mobile hardware configuration. The framework is able to continuously maintain the security status of the devices. With a high level of security status, users are permitted to access sensitive services and data. On the other hand, with the low level of security, users are required to re-authenticate before accessing sensitive service or data.

The relevance of faces in our daily lives is indisputable. We learn to recognize faces as newborns, and faces play a major role in interpersonal communication. The spectrum of computer vision research about face analysis includes, but is not limited to, face detection and facial attribute classification, which are the focus of this dissertation. The face is a primary biometric because by itself revels the subject's identity, while facial attributes (such as hair color and eye state) are soft biometrics because by themselves they do not reveal the subject's identity. In this dissertation, we proposed a real-time model for classifying 40 facial attributes, which preprocesses faces and then extracts 7 types of classical and deep features. These features were fused together to train 3 different classifiers. Our proposed model yielded 91.93% on the average accuracy outperforming 7 state-of-the-art models. We also developed a real-time model for classifying the states of human eyes and mouth (open/closed), and the presence/absence of eyeglasses in the wild. Our method begins by preprocessing a face by cropping the regions of interest (ROIs), and then describing them using RootSIFT features. These features were used to train a nonlinear support vector machine for each attribute. Our eye-state classifier achieved the top performance, while our mouth-state and glasses classifiers were tied as the top performers with deep learning classifiers. We also introduced a new facial attribute related to Middle Eastern headwear (called igal) along with its detector. Our proposed idea was to detect the igal using a linear multiscale SVM classifier with a HOG descriptor. Thereafter, false positives were discarded using dense SIFT filtering, bag-of-visual-words decomposition, and nonlinear SVM classification. Due to the similarity in real-life applications, we compared the igal detector with state-of-the-art face detectors, where the igal detector significantly outperformed the face detectors with the lowest false positives. We also fused the igal detector with a face detector to improve the detection performance. Face detection is the first process in any facial attribute classification pipeline. As a result, we reported a novel study that evaluates the robustness of current face detectors based on: (1) diffraction blur, (2) image scale, and (3) the IoU classification threshold. This study would enable users to pick the robust face detector for their intended applications.
Doctor of Philosophy
The relevance of faces in our daily lives is indisputable. We learn to recognize faces as newborns, and faces play a major role in interpersonal communication. Faces probably represent the most accurate biometric trait in our daily interactions. Thereby, it is not singular that so much effort from computer vision researchers have been invested in the analysis of faces. The automatic detection and analysis of faces within images has therefore received much attention in recent years. The spectrum of computer vision research about face analysis includes, but is not limited to, face detection and facial attribute classification, which are the focus of this dissertation. The face is a primary biometric because by itself revels the subject's identity, while facial attributes (such as hair color and eye state) are soft biometrics because by themselves they do not reveal the subject's identity. Soft biometrics have many uses in the field of biometrics such as (1) they can be utilized in a fusion framework to strengthen the performance of a primary biometric system. For example, fusing a face with voice accent information can boost the performance of the face recognition. (2) They also can be used to create qualitative descriptions about a person, such as being an "old bald male wearing a necktie and eyeglasses." Face detection and facial attribute classification are not easy problems because of many factors, such as image orientation, pose variation, clutter, facial expressions, occlusion, and illumination, among others. In this dissertation, we introduced novel techniques to classify more than 40 facial attributes in real-time. Our techniques followed the general facial attribute classification pipeline, which begins by detecting a face and ends by classifying facial attributes. We also introduced a new facial attribute related to Middle Eastern headwear along with its detector. The new facial attribute were fused with a face detector to improve the detection performance. In addition, we proposed a new method to evaluate the robustness of face detection, which is the first process in the facial attribute classification pipeline. Detecting the states of human facial attributes in real time is highly desired by many applications. For example, the real-time detection of a driver's eye state (open/closed) can prevent severe accidents. These systems are usually called driver drowsiness detection systems. For classifying 40 facial attributes, we proposed a real-time model that preprocesses faces by localizing facial landmarks to normalize faces, and then crop them based on the intended attribute. The face was cropped only if the intended attribute is inside the face region. After that, 7 types of classical and deep features were extracted from the preprocessed faces. Lastly, these 7 types of feature sets were fused together to train three different classifiers. Our proposed model yielded 91.93% on the average accuracy outperforming 7 state-of-the-art models. It also achieved state-of-the-art performance in classifying 14 out of 40 attributes. We also developed a real-time model that classifies the states of three human facial attributes: (1) eyes (open/closed), (2) mouth (open/closed), and (3) eyeglasses (present/absent). Our proposed method consisted of six main steps: (1) In the beginning, we detected the human face. (2) Then we extracted the facial landmarks. (3) Thereafter, we normalized the face, based on the eye location, to the full frontal view. (4) We then extracted the regions of interest (i.e., the regions of the mouth, left eye, right eye, and eyeglasses). (5) We extracted low-level features from each region and then described them. (6) Finally, we learned a binary classifier for each attribute to classify it using the extracted features. Our developed model achieved 30 FPS with a CPU-only implementation, and our eye-state classifier achieved the top performance, while our mouth-state and glasses classifiers were tied as the top performers with deep learning classifiers. We also introduced a new facial attribute related to Middle Eastern headwear along with its detector. After that, we fused it with a face detector to improve the detection performance. The traditional Middle Eastern headwear that men usually wear consists of two parts: (1) the shemagh or keffiyeh, which is a scarf that covers the head and usually has checkered and pure white patterns, and (2) the igal, which is a band or cord worn on top of the shemagh to hold it in place. The shemagh causes many unwanted effects on the face; for example, it usually occludes some parts of the face and adds dark shadows, especially near the eyes. These effects substantially degrade the performance of face detection. To improve the detection of people who wear the traditional Middle Eastern headwear, we developed a model that can be used as a head detector or combined with current face detectors to improve their performance. Our igal detector consists of two main steps: (1) learning a binary classifier to detect the igal and (2) refining the classier by removing false positives. Due to the similarity in real-life applications, we compared the igal detector with state-of-the-art face detectors, where the igal detector significantly outperformed the face detectors with the lowest false positives. We also fused the igal detector with a face detector to improve the detection performance. Face detection is the first process in any facial attribute classification pipeline. As a result, we reported a novel study that evaluates the robustness of current face detectors based on: (1) diffraction blur, (2) image scale, and (3) the IoU classification threshold. This study would enable users to pick the robust face detector for their intended applications. Biometric systems that use face detection suffer from huge performance fluctuation. For example, users of biometric surveillance systems that utilize face detection sometimes notice that state-of-the-art face detectors do not show good performance compared with outdated detectors. Although state-of-the-art face detectors are designed to work in the wild (i.e., no need to retrain, revalidate, and retest), they still heavily depend on the datasets they originally trained on. This condition in turn leads to variation in the detectors' performance when they are applied on a different dataset or environment. To overcome this problem, we developed a novel optics-based blur simulator that automatically introduces the diffraction blur at different image scales/magnifications. Then we evaluated different face detectors on the output images using different IoU thresholds. Users, in the beginning, choose their own values for these three settings and then run our model to produce the efficient face detector under the selected settings. That means our proposed model would enable users of biometric systems to pick the efficient face detector based on their system setup. Our results showed that sometimes outdated face detectors outperform state-of-the-art ones under certain settings and vice versa.

The identification of laboratory mice has been an important issue in pharmaceutical applications ever since tests have been performed on animals. As biometric identification has become an increasingly important issue over the past decade, attempts are underway to replace traditional identification methods, which are mostly invasive and limited in code space. This thesis discusses a project that aims at identifying mice by biometrically examining the blood vessel patterns in their ears.

In the proposed algorithm, firstly, the blood vessel structure within the obtained images got enhanced before segmenting the image in blood vessel and non-blood vessel portions. In the next step a sufficient amount of unique feature points got extracted from the segmented image. The obtained feature points were afterwards used for the actual identification procedure.

Out of 20 mice, 18 could be identified successfully using the proposed algorithm.

The work presented within this document details the development of a novel gait verification system suitable for a variety of applications such as human motion studies, medical analysis and security situations. The human gait is a spatio temporal process involving the coordination and interaction between the nervous, skeletal and muscular systems. Due to inherent variations in the limb lengths, muscle strengths and body mass gait is inherently individual. To develop a suitable feature extraction process a virtual gait laboratory was developed. The virtual laboratory contains virtual character templates articulated with a 32 bone skeleton system using motion capture data. Data was extracted from the character as a series of X, Y and Z translations for pro cessing. The virtual laboratory allows the testing of data extraction processes without the need for direct testing on human subjects. Feature extraction was performed using Principal Component Analysis (PCA). PCA allows data to be compressed and describes as a series of principal scores (PC) containing the weightings of the data. Feature extraction was performed on human subjects with the motions applied to a skeletal system containing individual physical dimensions. A second set of features was created by applying the motions to a single skeletal system. This removed the interpersonal variations from the dataset to explore the difference in classification when these variables have been removed. Overall generic motions are present within the first PC score. Higher PC scores contain unique motion characteristics suitable for classification of the subject s within a database. To verify a subject within the database Linear Discriminant Analysis (LOA) was performed. LOA projects data as a linear combination of features using a t raining data set of known outcomes. A subsequent sample can then be projected into the linear space for classification and verification of the subject within the database.

Approved for public release; distribution is unlimited
This thesis investigates the ties between biometrics and state security by analyzing biometric identification and screening programs, their structural elements, and ultimately their effectiveness. Although biometric identification is rapidly becoming an international norm, quantitative assessments of biometric identification programs within the larger context of state and international security are non-existent. This thesis discusses the idea of identity, defines the identity problem, addresses identity's role in state security, and addresses how biometric identification contributes to this end. Individual characteristics of the most prominently used biometric identifiers are discussed in detail (face, fingerprint, and iris), as well as the overall concept of biometric identification. The ICAO e-Passport program and the U.S. specific screening functions are presented to illuminate how biometric identifiers are used in practical applications. These programs, in turn, serve as the basis for the investigation of the effectiveness of biometric identification as it pertains to state security, focusing first on U.S. immigration and then on the broader context of international terrorism. Biometric identification has been largely credited with producing tangible security gains. This thesis seeks to tie a quantitative measure to that assertion and generate future discussion about the merits of biometrically based identification and screening.

We have designed and demonstrated a new class of medical navigation methods that use the fingerprint-like biometrically distinct ultrasound echo patterns produced by different locations in tissue. As an example of this new biometric navigation approach, we have constructed and tested a system that uses ultrasound data to achieve prospective motion compensation in MRI, especially for respiratory motion during interventional MRI procedures in moving organs such as the liver. The ultrasound measurements are collated with geometrical information from MRI during a training stage to form a mapping table that relates ultrasound measurements to positions. During prospective correction, the system makes frequent ultrasound measurements and uses the map to determine the corresponding position. Results in motorized linear motion phantoms and freely breathing animals indicate that the system performs well. Apparent motion is reduced by up to 97.8%, and motion artifacts are reduced or eliminated in 2D Spoiled Gradient-Echo images. The motion compensation is sufficient to permit MRI thermometry of focused ultrasound heating during respiratory-like motion, with results similar to those obtained in the absence of motion. This new technique may have applications for MRI thermometry and other dynamic imaging in the abdomen during free breathing. We have also extended this technique to situations in which external position information during training is unavailable or incomplete, by extending the concept of Simultaneous Localization and Mapping to include determining the topology of a dense motion path through a gaussian random field. In the course of these investigations, we have also developed modified forms of referenceless MRI thermometry and Kalman filtering, specially adapted to optimize accuracy under our experimental conditions.

The iris biometric is a well-established technology which is already in use in several nation-scale applications and it is still an active research area with several unsolved problems. This work focuses on three key problems in iris biometrics namely: segmentation, protection and cross-matching. Three novel methods in each of these areas are proposed and analyzed thoroughly. In terms of iris segmentation, a novel iris segmentation method is designed based on a fusion of an expanding and a shrinking active contour by integrating a new pressure force within the Gradient Vector Flow (GVF) active contour model. In addition, a new method for closed eye detection is proposed. The experimental results on the CASIA V4, MMU2, UBIRIS V1 and UBIRIS V2 databases show that the proposed method achieves state-of-theart results in terms of segmentation accuracy and recognition performance while being computationally more efficient. In this context, improvements by 60.5%, 42% and 48.7% are achieved in segmentation accuracy for the CASIA V4, MMU2 and UBIRIS V1 databases, respectively. For the UBIRIS V2 database, a superior time reduction is reported (85.7%) while maintaining a similar accuracy. Similarly, considerable time improvements by 63.8%, 56.6% and 29.3% are achieved for the CASIA V4, MMU2 and UBIRIS V1 databases, respectively. With respect to iris biometric protection, a novel security architecture is designed to protect the integrity of iris images and templates using watermarking and Visual Cryptography (VC). Firstly, for protecting the iris image, text which carries personal information is embedded in the middle band frequency region of the iris image using a novel watermarking algorithm that randomly interchanges multiple middle band pairs of the Discrete Cosine Transform (DCT). Secondly, for iris template protection, VC is utilized to protect the iii iris template. In addition, the integrity of the stored template in the biometric smart card is guaranteed by using the hash signatures. The proposed method has a minimal effect on the iris recognition performance of only 3.6% and 4.9% for the CASIA V4 and UBIRIS V1 databases, respectively. In addition, the VC scheme is designed to be readily applied to protect any biometric binary template without any degradation to the recognition performance with a complexity of only O(N). As for cross-spectral matching, a framework is designed which is capable of matching iris images in different lighting conditions. The first method is designed to work with registered iris images where the key idea is to synthesize the corresponding Near Infra-Red (NIR) images from the Visible Light (VL) images using an Artificial Neural Network (ANN) while the second method is capable of working with unregistered iris images based on integrating the Gabor filter with different photometric normalization models and descriptors along with decision level fusion to achieve the cross-spectral matching. A significant improvement by 79.3% in cross-spectral matching performance is attained for the UTIRIS database. As for the PolyU database, the proposed verification method achieved an improvement by 83.9% in terms of NIR vs Red channel matching which confirms the efficiency of the proposed method. In summary, the most important open issues in exploiting the iris biometric are presented and novel methods to address these problems are proposed. Hence, this work will help to establish a more robust iris recognition system due to the development of an accurate segmentation method working for iris images taken under both the VL and NIR. In addition, the proposed protection scheme paves the way for a secure iris images and templates storage. Moreover, the proposed framework for cross-spectral matching will help to employ the iris biometric in several security applications such as surveillance at-a-distance and automated watch-list identification.

Dans son contexte spécifique, le terme « biométrie » est souvent associé à l'étude des caractéristiques physiques et comportementales des individus afin de parvenir à leur identification ou à leur vérification. Ainsi, le travail développé dans cette thèse nous a conduit à proposer un algorithme d'identification robuste, en considérant les caractéristiques intrinsèques des phalanges de la main. Considérée comme une biométrie cachée, cette nouvelle approche peut s'avérer intéressante, notamment lorsqu'il est question d'assurer un niveau de sécurité élevé, robuste aux différentes attaques qu'un système biométrique doit contrer. La base des techniques proposées requière trois phases, à savoir: (1) la segmentation des phalanges, (2) l'extraction de leurs caractéristiques par la génération d'une empreinte, appelée « Phalange-Code » et (3) l'identification basée sur la méthode du 1-plus proche voisin ou la vérification basée sur une métrique de similarité. Ces algorithmes opèrent sur des niveaux hiérarchiques permettant l'extraction de certains paramètres, invariants à des transformations géométriques telles que l'orientation et la translation. De plus, nous avons considéré des techniques robustes au bruit, pouvant opérer à différentes résolutions d'images. Plus précisément, nous avons élaboré trois approches de reconnaissance biométrique : la première approche utilise l'information spectrale des contours des phalanges de la main comme signature individuelle, alors que la deuxième approche nécessite l'utilisation des caractéristiques géométriques et morphologiques des phalanges (i.e. surface, périmètre, longueur, largeur, capacité). Enfin, la troisième approche requière la génération d'un nouveau rapport de vraisemblance entre les phalanges, utilisant la théorie de probabilités géométriques. En second lieu, la construction d'une base de données avec la plus faible dose de rayonnement a été l'un des grands défis de notre étude. Nous avons donc procédé par la collecte de 403 images radiographiques de la main, acquises en utilisant la machine Apollo EZ X-Ray. Ces images sont issues de 115 adultes volontaires (hommes et femmes), non pathologiques. L'âge moyen étant de 27.2 ans et l'écart-type est de 8.5. La base de données ainsi construite intègre des images de la main droite et gauche, acquises à des positions différentes et en considérant des résolutions différentes et des doses de rayonnement différentes (i.e. réduction jusqu'à 98 % de la dose standard recommandée par les radiologues « 1 µSv »).Nos expériences montrent que les individus peuvent être distingués par les caractéristiques de leurs phalanges, que ce soit celles de la main droite ou celles de la main gauche. Cette distinction est également valable pour le genre des individus (homme/femme). L'étude menée a montré que l'approche utilisant l'information spectrale des contours des phalanges permet une identification par seulement trois phalanges, à un taux EER (Equal Error Rate) inférieur à 0.24 %. Par ailleurs, il a été constaté « de manière surprenante » que la technique fondée sur les rapports de vraisemblance entre les phalanges permet d'atteindre un taux d'identification de 100 % et un taux d'EER de 0.37 %, avec une seule phalange. Hormis l'aspect identification/authentification, notre étude s'est penchée sur l'optimisation de la dose de rayonnement permettant une identification saine des individus. Ainsi, il a été démontré qu'il était possible d'acquérir plus de 12500/an d'images radiographiques de la main, sans pour autant dépasser le seuil administratif de 0.25 mSv
In its specific context, the term "biometrics" is often associated with the study of the physical and behavioral of individual's characteristics to achieve their identification or verification. Thus, the work developed in this thesis has led us to suggest a robust identification algorithm, taking into account the intrinsic characteristics of the hand phalanges. Considered as hidden biometrics, this new approach can be of high interest, particularly when it comes to ensure a high level of security, robust to various attacks that a biometric system must address. The basis of the proposed techniques requires three phases, namely: (1) the segmentation of the phalanges (2) extracting their characteristics by generating an imprint, called "Phalange-Code" and (3) the identification based on the method of 1-nearest neighbor or the verification based on a similarity metric. This algorithm operates on hierarchical levels allowing the extraction of certain parameters invariant to geometric transformations such as image orientation and translation. Furthermore, the considered algorithm is particularly robust to noise, and can function at different resolutions of images. Thus, we developed three approaches to biometric recognition: the first approach produces individual signature from the spectral information of the contours issued from the hand phalanges, whereas the second approach requires the use of geometric and morphological characteristics of the phalanges (i.e. surface, perimeter, length, width, and capacity). Finally, the third approach requires the generation of a new likelihood ratio between the phalanges, using the geometric probability theory. Furthermore, the construction of a database with the lowest radiation dose was one of the great challenges of our study. We therefore proceeded with the collection of 403 x-ray images of the hand, acquired using the Apollo EZ X-Ray machine. These images are from 115 non-pathological volunteering adult (men and women). The average age is 27.2 years and the standard deviation is 8.5. Thus, the constructed database incorporates images of the right and left hands, acquired at different positions and by considering different resolutions and different radiation doses (i.e. reduced till 98% of the standard dose recommended by radiologists "1 µSv").Our experiments show that individuals can be distinguished by the characteristics of their phalanges, whether those of the right hand or the left hand. This distinction also applies to the kind of individuals (male/female). The study has demonstrated that the approach using the spectral information of the phalanges' contours allows identification by only three phalanges, with an EER (Equal Error Rate) lower than 0.24 %. Furthermore, it was found “Surprisingly” that the technique based on the likelihood ratio between phalanges reaches an identification rate of 100% and an EER of 0.37% with a single phalanx. Apart from the identification/authentication aspect, our study focused on the optimization of the radiation dose in order to offer safe identification of individuals. Thus, it has been shown that it was possible to acquire more than 12,500/year radiographic hand images, without exceeding the administrative control of 0.25 mSv

A proposal for a person authentication system, which localizes facial landmarks and extracts biometrical features for face authentication, is presented in this thesis. An efficient algorithm for eye localization and biometrical feature extraction and person identification is developed by using Gabor filters. In the eye localization part, we build artificial average eye models for eye location. In the person identification part, we construct databases of biometrical features around the eye area of clients and, for authentication, Schwartz inequality and the sum square error (SSE) are used.

This project is implemented in the ‘Matlab’ programming language, on a personal computer system, and experimental results on the proposed system are presented.

The recent growth in CCTV systems and the challenges of automatically identifying humans under the adverse visual conditions of surveillance have increased the interest in soft biometrics, which are physical and behavioural attributes that are used to semantically describe people. Soft biometrics enable human identification under the challenging conditions of surveillance where it is impossible to acquire traditional biometrics such as iris and fingerprint. The existing work on facial soft biometrics is focused on categorical attributes, while comparative attributes have received very little attention, although they have demonstrated a better accuracy. Thus, it is still unknown whether comparative soft biometrics can scale to large and more realistic databases. Also, the automatic retrieval of comparative facial soft biometrics from images needs to be investigated. The purpose of this thesis is to explore human identification and verification in large and realistic databases via comparative facial soft biometrics using the Labelled Faces in the Wild (LFW) database. A novel set of comparative facial soft biometrics is introduced, and a thorough analysis that assesses attribute significance and discriminative power is presented. Also, a set of identification and verification experiments was conducted to evaluate the comparative facial soft biometrics. Moreover, this thesis proposes MIURank, a novel fully unsupervised ranking algorithm that is based on mutual information. The experiments demonstrate that a correct match can be found in the top 71 retrieved subjects from a database of 4038 subjects by comparing an unknown subject to ten subjects only. Additionally, the experiments reveal that face retrieval by verbal descriptions in a database of images can yield a correct match in the top 15 retrieved subjects from a database of 430 subjects. Furthermore, the performance analysis of the MIURank algorithm shows that it can result in a ranking accuracy that is comparable to the maximum likelihood estimator of Bradley-Terry and the state-of-the-art SerialRank algorithm. By these analyses and developments, it is now possible not only to use human labels for recognition, but also to derive them by computer vision.

Computer security is a growing field in the IT industry. One of the important aspects of the computer security is authentication. Using passwords (something you know) is one of the most common ways of authentications. But passwords have proven to provide weak level of security as they can be easily compromised. Some other ways of authenticating a user are using physical tokens, (something you possess) and biometrics, (something you are). Using any one of these techniques to secure a system always has its own set of threats. One way to make sure a system is secure is to use multiple factors to authenticate. One of the ways to use multiple factors is to use all the three factors of authentication, something you possess, something you are and something you know. This thesis discusses about different ways of authentication and implements a system using three factor authentication. It takes many security aspects of the system into consideration while implementing it, to make it secure.

Recently, soft biometrics has emerged as a novel attribute-based person description for identification. It is likely that soft biometrics can be deployed where other biometrics cannot, and have stronger invariance properties than traditional vision-based biometrics, such as invariance to illumination and contrast. Previously, a variety of soft body and face biometrics have been used for identifying people and have increasingly garnered more research interest and are often considered as major cues for identity, especially in the absence of valid traditional hard biometrics, as in surveillance. Describing a person by their clothing properties is a natural task performed by people. As yet, clothing descriptions have attracted little attention for biometric purposes as it has been considered unlikely to be a potential cue to identity. There has been some usage of clothing attributes to augment biometric description, but a detailed description has yet to be used. In everyday life, several cases and incidents arise highlighting the usefulness and capability of information deduced from clothing regarding identity. Clothing is inherently more effective for short-term identification, since people can change clothes. This thesis introduces semantic clothing attributes as a new form of soft biometrics. The usability and efficacy of a novel set of proposed soft clothing traits is explored, showing how they can be exploited for human identification and re-identification purposes. Furthermore, the viability of these traits is investigated in correctly retrieving a subject of interest, given a verbal description of their clothing. The capability of clothing information is further examined in more realistic scenarios offering viewpoint invariant subject retrieval. Although clothing traits can be naturally described or compared by humans for operable and successful use, it is desirable to exploit computer-vision to enrich clothing descriptions with more objective and discriminative information. This allows automatic extraction and semantic description and comparison of visually detectable clothing traits in a manner similar to recognition by eyewitness statements. This thesis proposes further a novel set of automatic clothing attributes, described using small groups of high-level semantic labels, and automatically extracted using computer-vision techniques. In this way, we can explore the capability of clothing attributes inferred by human vis-a-vis those which are inferred automatically by computer-vision. Extended analysis of clothing information is conducted. Human identification and retrieval are achieved, evaluated, and compared using different proposed forms of soft clothing biometrics in addition and in isolation. The experimental results of identification and retrieval highlight clothing attributes as a potentially valuable addition to the field of soft biometrics.

Recognising humans by the way they walk has attracted a significant interest in recent years due to its potential use in a number of applications such as automated visual surveillance. Technologies utilising gait biometrics have the potential to provide safer society and improve quality of life. However, automated gait recognition is a very challenging research problem and some fundamental issues remain unsolved. At the moment, gait recognition performs well only when samples acquired in similar conditions are matched. An operational automated gait recognition system does not yet exist. The primary aim of the research presented in this thesis is to understand the main challenges associated with deployment of gait recognition and to propose novel solutions to some of the most fundamental issues. There has been lack of understanding of the effect of some subject dependentcovariates on gait recognition performance. We have proposed a novel dataset that allows analyses of various covariates in a principled manner. The results of thedatabase evaluation revealed that elapsed time does not affect recognition in the short to medium term, contrary to what other studies have concluded. The analyses show how other factors related to the subject affect recognition performance. Only few gait recognition approaches have been validated in real world conditions. We have collected a new dataset at two realistic locations. Using the database we have shown that there are many environment related factors that can affect performance. The quality of silhouettes has been identified as one of the most important issues for translating gait recognition research to the ‘real-world’. The existing quality algorithms proved insufficient and therefore we extended quality metrics and proposed new ways of improving signature quality and therefore performance. A new fully working automated system has been implemented. Experiments using the system in ‘real-world’ conditions have revealed additional challenges not present when analysing datasets of fixed size. In conclusion, the research has investigated many of the factors that affect current gait recognition algorithms and has presented novel approaches of dealing with some of the most important issues related to translating gait recognition to real-world environments.