Дисертації з теми "Multibiometric"

This thesis has aimed to explore an application of Multibiometrics to secured wireless communications. The medium of study for this purpose included Wi-Fi, 3G, and WiMAX, over which simulations and experimental studies were carried out to assess the performance. In specific, restriction of access to authorized users only is provided by a technique referred to hereafter as multibiometric cryptosystem. In brief, the system is built upon a complete challenge/response methodology in order to obtain a high level of security on the basis of user identification by fingerprint and further confirmation by verification of the user through text-dependent speaker recognition. First is the enrolment phase by which the database of watermarked fingerprints with memorable texts along with the voice features, based on the same texts, is created by sending them to the server through wireless channel. Later is the verification stage at which claimed users, ones who claim are genuine, are verified against the database, and it consists of five steps. Initially faced by the identification level, one is asked to first present one’s fingerprint and a memorable word, former is watermarked into latter, in order for system to authenticate the fingerprint and verify the validity of it by retrieving the challenge for accepted user. The following three steps then involve speaker recognition including the user responding to the challenge by text-dependent voice, server authenticating the response, and finally server accepting/rejecting the user. In order to implement fingerprint watermarking, i.e. incorporating the memorable word as a watermark message into the fingerprint image, an algorithm of five steps has been developed. The first three novel steps having to do with the fingerprint image enhancement (CLAHE with 'Clip Limit', standard deviation analysis and sliding neighborhood) have been followed with further two steps for embedding, and extracting the watermark into the enhanced fingerprint image utilising Discrete Wavelet Transform (DWT). In the speaker recognition stage, the limitations of this technique in wireless communication have been addressed by sending voice feature (cepstral coefficients) instead of raw sample. This scheme is to reap the advantages of reducing the transmission time and dependency of the data on communication channel, together with no loss of packet. Finally, the obtained results have verified the claims.

Thesis (Ph. D.)--Michigan State University. Dept. of Computer Science and Engineering, 2008.
Title from PDF t.p. (viewed on Mar. 30, 2009) Includes bibliographical references (p. 210-228). Also issued in print.

E-learning has grown to such an extent that paper-based testing is being replaced by computer-based testing otherwise known as e-exams. Because these e-exams can be delivered outside of the traditional proctored environment, additional authentication measures must be employed in order to offer similar authentication assurance as found in proctored, paper-based testing. This dissertation addressed the need for valid authentication in e-learning systems, in e-examinations in particular, and especially in professional certification e-examinations. Furthermore, this dissertation proposed a more robust method for learner authentication during e-examination taking. Finally, this dissertation extended e-learning research by comparing e-examination scores and durations of three separate groups of exam takers using different authentication methods: Online Using Username/Password (OLUP), In-Testing Center (ITC), and Online with Multibiometrics (OLMB) to better understand the role as well as the possible effect of continuous and dynamic multibiometric authentication on professional certification e-examination scores and durations. The sample used in this study was based on participants who were all professional members of a technology professional certification organization. The methodology used to collect data was a posttest only, multiple, non-equivalent groups quasi-experiment, where age, gender, and Information Technology Proficiency (ITP) were also recorded. The analyses performed in this study included pre-analysis data screening, reliability analyses for each instrument used, and the main analysis to address each hypothesis. Group affiliation, i.e. type of authentication methods, was found to have no significant effect on differences among exam scores and durations. While there was a clear path of increased mean e-examination score as authentication method was relaxed, it was evident from the analysis that these were not significant differences. Age was found to have a significant effect on exam scores where younger participants were found to have higher exam scores and lower exam durations than older participants. Gender was not found to have a significant effect on exam scores nor durations. ITP was found to have a significant effect on exam scores and durations where greater scores with the ITP instrument indicated greater exam scores and lower exam durations. This study's results can help organizations better understand the role, possible effect, and potential application of continuous and dynamic multibiometric authentication as a justifiable approach when compared with the more common authentication approach of User Identifier (UID) and password, both in professional certification e-examinations as well as in an online environment.

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.

In the digital and interconnected world we live in, establishing the identity of any individual is a pressing need. Home banking, on line shopping, and social care web sites are only few examples of services where proof of identity is fundamental. Such a process can be based on "what you know" (i.g. a password), on"what you posses" (i.g. the key of a house or an ID card) or on "what you are"(ID-based, i.g. biometrics). In this thesis we focus on biometrics. Biometric recognition, or simply biometrics, refers to ``the automated recognition of individuals based on behavioral and biological characteristics'' (ISO/IEC JTC1 SC37). This method of recognition has the advantage that it does not need the memorization of any password or the possess of any token, at the same time, however, biometrics cannot be changed if compromised in any way, hence calling for the adoption of suitable protection mechanisms. In this thesis we study the development of privacy preserving protocols for biometric recognition. This is a new research field for which a number of solutions have been proposed in recent years. For efficiency reasons, the majority of those solutions are secure only against a passive adversary, that is an adversary that does not deviate from the protocol, yet tries to infer as much information as possible from the data exchanged during the protocol. On the contrary, in this thesis we look for protocols which are secure against active adversaries, that is adversaries that deliberately and arbitrarily deviate from the recognition protocol. Specifically, we propose two possible solutions using signal processing in the encrypted domain's tools. First we use a cryptographic scheme belonging to the somewhat homomorphic scheme's family and we propose both an identification and an authentication non-interactive scheme. The first protocol focuses on a one-to-many recognition task: the biometric probe of a specific individual is compared with all the probes contained in a database looking for a positive match. The second protocol, instead, considers a one to one comparison. The new probe of an enrolled individual is compared with the probe of the same individual stored during the enrollment phase. As a second contribution, we propose SEMBA: a protocol secure against active adversary for multibiometric recognition. In this case we look for a trade-off between efficiency and accuracy by combining information from two biometric traits instead of only one. The protocol relies on SPDZ, a new framework proposed by Damgård et al. which is secure also in the presence of an active adversary.

A combinação de fontes biométricas não redundantes da multibiometria supera a precisão de cada fonte individual (monobiometria). Além do mais, dois problemas em biometria, ruído e ataques de usurpadores, podem ser minimizados pelo uso de múltiplos sensores e biometria multimodal. Entretanto, se as similaridades estão em todos traços biométricos, como em gêmeos monozigotos (MZ), o processamento de múltiplas fontes não melhora a performance. Para distinguir extrema similitude, influências epigenéticas e ambientais são mais importantes do que o DNA herdado. Esta tese examina a plasticidade fenotípica na assimetria humana como uma ferramenta para melhorar a multibiometria. A técnica de Processamento Bilateral (PB) é introduzida para analisar discordâncias em lados esquerdo e direito dos traços biométricos. PB foi testado com imagens de espectro visível e infravermelho usando Correlação Cruzada, Wavelets e Redes Neurais Artificiais. Os traços selecionados foram dentes, orelhas, íris, impressões digitais, narinas e bochechas. PB acústico também foi implementado para avaliação da assimetria vibracional durante sons vocálicos e comparado a um sistema reconhecedor de locutores com parametrização via MFCC (Mel Frequency Cepstral Coefficients) e classificado por Quantização Vetorial. Para o PB de imagens e acústico foram coletadas 20 amostras por traço biométrico durante um ano de nove irmãos masculinos adultos. Com propósito de teste, as biometrias esquerdas foram impostoras às biometrias direitas do mesmo indivíduo e vice-versa, o que levou a 18 entidades serem identificadas por traço biométrico. Resultados alcançaram identificação total em todas biometrias tratadas com PB, comparado a um máximo de 44% de identificação correta sem PB. Esta tese conclui que peculiaridades bilaterais melhoram a performance multibiométrica e podem complementar qualquer abordagem de reconhecimento.
Combination of non-redundant biometric sources in multibiometrics overcomes individual source accuracy (monobiometrics). Moreover, two problems in biometrics, noise and impostor attacks, can be minimized by the use of multi-sensor, multi-modal biometrics. However, if similarities are in all traits, as in monozygotic twins (MZ), multiple source processing does not improve performance. To distinguish extreme similitude, epigenetic and environmental influences are more important than DNA inherited. This thesis examines phenotypic plasticity in human asymmetry as a tool to ameliorate multibiometrics. Bilateral Processing (BP) technique is introduced to analyze discordances in left and right trait sides. BP was tested in visible and infrared spectrum images using Cross-Correlation, Wavelets and Artificial Neural Networks. Selected traits were teeth, ears, irises, fingerprints, nostrils and cheeks. Acoustic BP was also implemented for vibration asymmetry evaluation during voiced sounds and compared to a speaker recognition system parameterized via MFCC (Mel Frequency Cepstral Coefficients) and classified by Vector Quantization. Image and acoustic BP gathered 20 samples per biometric trait during one year from nine adult male brothers. For test purposes, left biometrics was impostor to right biometrics from the same individual and vice-versa, which led to 18 entities to be identified per trait. Results achieved total identification in all biometrics treated with BP, compared to maximum 44% of correct identification without BP. This study concludes that bilateral peculiarities improve multibiometric performance and can complement any recognition approach.

Made available in DSpace on 2014-06-11T19:29:40Z (GMT). No. of bitstreams: 0 Previous issue date: 2008-07-04Bitstream added on 2014-06-13T19:38:57Z : No. of bitstreams: 1 falguera_fps_me_sjrp.pdf: 3832818 bytes, checksum: 1ca4e6b68ff66693475c6e5caed03e88 (MD5)
Biometria refere-se ao uso de características físicas (impressões digitais, íris, retina) ou comportamentais (assinatura, voz) para a identificação humana. As impressões digitais são formadas por cristas e minúcias. As cristas são linhas distribuídas paralelamente com uma orientação e um espaçamento característico e as minúcias representam os vários modos pelos quais uma crista pode se tornar descontínua. Graças a sua universalidade, unicidade e permanência, as impressões digitais tornaram-se as características biométricas mais amplamente utilizadas. Entretanto, considerar o reconhecimento automático de impressões digitais um problema totalmente resolvido é um erro muito comum. Nenhum sistema de reconhecimento de impressões digitais proposto até hoje é infalível, nenhum garante taxas de erro nulas. Imagens de baixa qualidade e com pequena área de sobreposição entre a imagem template e a imagem de consulta ainda representam um desafio para os métodos de reconhecimento de impressões digitais mais utilizados, os métodos baseados no casamento de pontos de minúcias. Uma das maneiras de superar as limitações e melhorar a acurácia de um sistema biométrico é o uso da multibiometria, isto é, a combinação de diferentes tipos de informação em um sistema de reconhecimento biométrico. Neste contexto, esta dissertação de mestrado objetiva aprimorar a acurácia dos sistemas de reconhecimento de impressões digitais por meio da fusão de métodos baseados em minúcias e em cristas. Para tanto, foram implementadas técnicas de fusão no nível de pontuação, classificação e decisão. No nível de pontuação, a fusão propiciou uma redução na taxa de erro igual (EER) de 42,53% em relação ao método mais preciso. Para o nível de classificação, a fusão significou um aumento de 75% na taxa de recuperação correta...
Biometrics refers to the use of physical (fingerprints, iris, retina) or behavioral (signature, voice) characteristics to determine the identity of a person. Fingerprints are formed by ridges and minutiae. The ridges are lines distributed in parallel with an orientation and a characteristic spacing and the minutiae represent the several ways a ridge can become discontinued. As to its universality, uniqueness and permanence, the fingerprints became the most widely used biometric characteristic. However, it is a common mistake to consider the automatic fingerprint recognition as a totally solved problem. No fingerprint recognition system proposed until now is infallible, none of them guarantee null error rates. Poor quality images and when just a small area of overlap between the template and the query images exists are still a complex challenge to the most used fingerprint recognition methods, the methods based on minutiae points matching. One of the possibilities to overcome the limitations and improve the accuracy of a biometric system is the use of multibiometrics, the combination of different kinds of information in a biometric system. In this context, this master thesis aims to improve the accuracy of fingerprint recognition systems through the fusion of minutiae based and ridge based methods. To achieve this, fusion techniques on score, rank and decision levels were implemented. For the score level, the fusion lead to a reduction of the Equal Error Rate to 42.53% compared to the most precise method. For the rank level, the fusion meant an increase of 75% in the Correct Retrieval Rate. And, in the decision level fusion the Recognition Rate changed from 99.25% to 99.75%. The results have demonstrated that the fusion of minutiae based and ridge based methods can represent a significant accuracy improvement for the fingerprint recognition systems.

Multimodal biometric systems have been widely applied in many real-world applications due to its ability to deal with a number of significant limitations of unimodal biometric systems, including sensitivity to noise, population coverage, intra-class variability, non-universality, and vulnerability to spoofing. This PhD thesis is focused on the combination of both the face and the left and right irises, in a unified hybrid multimodal biometric identification system using different fusion approaches at the score and rank level. Firstly, the facial features are extracted using a novel multimodal local feature extraction approach, termed as the Curvelet-Fractal approach, which based on merging the advantages of the Curvelet transform with Fractal dimension. Secondly, a novel framework based on merging the advantages of the local handcrafted feature descriptors with the deep learning approaches is proposed, Multimodal Deep Face Recognition (MDFR) framework, to address the face recognition problem in unconstrained conditions. Thirdly, an efficient deep learning system is employed, termed as IrisConvNet, whose architecture is based on a combination of Convolutional Neural Network (CNN) and Softmax classifier to extract discriminative features from an iris image. Finally, The performance of the unimodal and multimodal systems has been evaluated by conducting a number of extensive experiments on large-scale unimodal databases: FERET, CAS-PEAL-R1, LFW, CASIA-Iris-V1, CASIA-Iris-V3 Interval, MMU1 and IITD and MMU1, and SDUMLA-HMT multimodal dataset. The results obtained have demonstrated the superiority of the proposed systems compared to the previous works by achieving new state-of-the-art recognition rates on all the employed datasets with less time required to recognize the person’s identity.Multimodal biometric systems have been widely applied in many real-world applications due to its ability to deal with a number of significant limitations of unimodal biometric systems, including sensitivity to noise, population coverage, intra-class variability, non-universality, and vulnerability to spoofing. This PhD thesis is focused on the combination of both the face and the left and right irises, in a unified hybrid multimodal biometric identification system using different fusion approaches at the score and rank level. Firstly, the facial features are extracted using a novel multimodal local feature extraction approach, termed as the Curvelet-Fractal approach, which based on merging the advantages of the Curvelet transform with Fractal dimension. Secondly, a novel framework based on merging the advantages of the local handcrafted feature descriptors with the deep learning approaches is proposed, Multimodal Deep Face Recognition (MDFR) framework, to address the face recognition problem in unconstrained conditions. Thirdly, an efficient deep learning system is employed, termed as IrisConvNet, whose architecture is based on a combination of Convolutional Neural Network (CNN) and Softmax classifier to extract discriminative features from an iris image. Finally, The performance of the unimodal and multimodal systems has been evaluated by conducting a number of extensive experiments on large-scale unimodal databases: FERET, CAS-PEAL-R1, LFW, CASIA-Iris-V1, CASIA-Iris-V3 Interval, MMU1 and IITD and MMU1, and SDUMLA-HMT multimodal dataset. The results obtained have demonstrated the superiority of the proposed systems compared to the previous works by achieving new state-of-the-art recognition rates on all the employed datasets with less time required to recognize the person’s identity.
Higher Committee for Education Development in Iraq

Penkiose disertacijos darbo dalyse nagrinėjamos asmens identifikavimo pagal pirštų atspaudus ir balsą problemos ir siūlomi jų sprendimai. Pirštų atspaudų požymių išskyrimo algoritmų kokybės įvertinimo problemą siūloma spręsti panaudojant sintezuotus pirštų atspaudus. Darbe siūlomos žinomo pirštų atpaudų sintezės algoritmo modifikacijos, kurios leidžia sukurti piršto atspaudo vaizdą su iš anksto nustatytomis charakteristikomis ir požymiais bei pagreitina sintezės procesą. Pirštų atspaudų požymių palyginimo problemos yra aptartos ir naujas palyginimo algoritmas yra siūlomas deformuotų pirštų palyginimui. Algoritmo kokybė yra įvertinta naudojant viešai prieinamas ir vidines duomenų bazes. Naujas asmens identifikavimo pagal balsą metodas remiantis tiesinės prognozės modelio grupinės delsos požymiais ir tų požymių palyginimo metrika kokybės prasme lenkia tradicinius asmens identifikavimo pagal balsą metodus. Pirštų ir balso įrašų nepriklausomumas yra irodytas ir asmens atpažinimas pagal balsą ir pirštų atspaudus kartu yra pasiūlytas siekiant išspręsti bendras biometrinių sistemų problemas.
This dissertation focuses on person identification problems and proposes solutions to overcome those problems. First part is about fingperprint feaures extraction algorithm performance evaluaiton. Modifications to a known synthesis algorithm are proposed to make it fast and suitable for performance evaluation. Matching of deformed fingerprints is discussed in the second part of the work. New fingerprint matching algorithm that uses local structures and does not perform fingerprint alignment is proposed to match deformed fingerprints. The use of group delay features of linear prediciton model for speaker identification is proposed in the third part of the work. New similarity metric that uses group delay features is described. It is demonstrated that automatic speaker identification system with proposed features and similarity metric outperforms traditional speaker identification systems. Multibiometrics using fingerprints and voice is adressed in the last part of the dissertation.

Made available in DSpace on 2014-06-11T19:29:40Z (GMT). No. of bitstreams: 0 Previous issue date: 2009-06-19Bitstream added on 2014-06-13T18:07:45Z : No. of bitstreams: 1 chiachia_g_me_sjrp.pdf: 1197775 bytes, checksum: a782f5b01605aa2a8b8bb080a56b3cad (MD5)
Coordenação de Aperfeiçoamento de Pessoal de Nível Superior (CAPES)
O reconhecimento facial é uma das principais formas de identificação humana. Apesar das pesquisas em reconhecimento facial automático terem crescido substancialmente ao longo dos últimos 35 anos, identificar pessoas a partir da face continua sendo um desafio para as áreas de Visão Computacional e Reconhecimento de Padrões. Em função dos cenários variarem desde a identificação a partir de fotografias até o reconhecimento baseado em vídeos sem nenhum tipo de controle ao serem gravados, os maiores desafios estão relacionados à independência contra diferentes tipos de iluminação, pose e expressão. O objetivo desta dissertação é propor técnicas que possam contribuir para a melhoria dos sistemas de reconhecimento facial. A primeira técnica endereça o problema da iluminação através da fusão dos espectros visível e infravermelho da face. Através desta abordagem, as taxas de reconhecimento foram melhoradas em 2.07% enquanto a taxa de erro igual (EER) foi reduzida em 45.47%. A segunda técnica trata do caso da extração e classificação de características faciais. Ela propõe um novo modelo para reconhecimento facial através do uso de características extraídas por Histogramas Census e de uma técnica de reconhecimento de padrões baseada em Máquinas de Vetores de Suporte (SVMs). Este outro grupo de experimentos nos possibilitou aumentar a precisão do reconhecimento no teste FERET fa/fb em 0.5%. Além destes resultados, algumas contribuições adicionais deste trabalho que merecem ser destacadas são a análise da dependência estatística entre classificadores de espectros diferentes e considerações sobre o comportamento de uma única C-SVC SVM para identificação de pessoas de forma eficaz.
Face recognition is one of the primary ways of human identification. Although researches on automated face recognition have broadly increased along the last 35 years, it remains a challenging task in the fields of Computer Vision and Pattern Recognition. As the scenarios varies from static and constrained photographs to uncontrolled video images, the challenging issues on automatic face recognition are usually related with variations in illumination, pose and expressions. The goal of this master thesis is to propose techniques for the improvement of face recognition systems. The first technique addresses the problem of illumination by fusing the visible and the infrared spectra of the face. With this approach the recognition rates were improved in 2.07% while the Equal Error Rate (EER) were reduced in 45.47%. The second technique addresses the issue of face features extraction and classification. It proposes a new framework for face recognition by using features extracted by Census Histograms and a pattern recognition technique based on Support Vector Machines (SVMs). This other group of experiments enabled us to increase the recognition accuracy in the FERET fa/fb test in 0.5%. Beyond these results, additional contributions of this work that deserve to be highlighted are the statistical dependency analysis between face recognition systems based on different spectra and a better comprehension about the behavior of a single C-SVC SVM to reliably predict faces identities.

Reliability of the performance of biometric identity verification systems remains a significant challenge. Individual biometric samples of the same person (identity class) are not identical at each presentation and performance degradation arises from intra-class variability and inter-class similarity. These limitations lead to false accepts and false rejects that are dependent. It is therefore difficult to reduce the rate of one type of error without increasing the other. The focus of this dissertation is to investigate a method based on classifier fusion techniques to better control the trade-off between the verification errors using text-dependent speaker verification as the test platform. A sequential classifier fusion architecture that integrates multi-instance and multisample fusion schemes is proposed. This fusion method enables a controlled trade-off between false alarms and false rejects. For statistically independent classifier decisions, analytical expressions for each type of verification error are derived using base classifier performances. As this assumption may not be always valid, these expressions are modified to incorporate the correlation between statistically dependent decisions from clients and impostors. The architecture is empirically evaluated by applying the proposed architecture for text dependent speaker verification using the Hidden Markov Model based digit dependent speaker models in each stage with multiple attempts for each digit utterance. The trade-off between the verification errors is controlled using the parameters, number of decision stages (instances) and the number of attempts at each decision stage (samples), fine-tuned on evaluation/tune set. The statistical validation of the derived expressions for error estimates is evaluated on test data. The performance of the sequential method is further demonstrated to depend on the order of the combination of digits (instances) and the nature of repetitive attempts (samples). The false rejection and false acceptance rates for proposed fusion are estimated using the base classifier performances, the variance in correlation between classifier decisions and the sequence of classifiers with favourable dependence selected using the 'Sequential Error Ratio' criteria. The error rates are better estimated by incorporating user-dependent (such as speaker-dependent thresholds and speaker-specific digit combinations) and class-dependent (such as clientimpostor dependent favourable combinations and class-error based threshold estimation) information. The proposed architecture is desirable in most of the speaker verification applications such as remote authentication, telephone and internet shopping applications. The tuning of parameters - the number of instances and samples - serve both the security and user convenience requirements of speaker-specific verification. The architecture investigated here is applicable to verification using other biometric modalities such as handwriting, fingerprints and key strokes.

Com o avanço da tecnologia, as estratégias tradicionais para identificação de pessoas se tornaram mais suscetíveis a falhas. De forma a superar essas dificuldades algumas abordagens vêm sendo propostas na literatura. Dentre estas abordagens destaca-se a Biometria. O campo da Biometria abarca uma grande variedade de tecnologias usadas para identificar ou verificar a identidade de uma pessoa por meio da mensuração e análise de aspectos físicos e/ou comportamentais do ser humano. Em função disso, a biometria tem um amplo campo de aplicações em sistemas que exigem uma identificação segura de seus usuários. Os sistemas biométricos mais populares são baseados em reconhecimento facial ou em impressões digitais. Entretanto, existem sistemas biométricos que utilizam a íris, varredura de retina, voz, geometria da mão e termogramas faciais. Atualmente, tem havido progresso significativo em reconhecimento automático de face em condições controladas. Em aplicações do mundo real, o reconhecimento facial sofre de uma série de problemas nos cenários não controlados. Esses problemas são devidos, principalmente, a diferentes variações faciais que podem mudar muito a aparência da face, incluindo variações de expressão, de iluminação, alterações da pose, assim como oclusões parciais. Em comparação com o grande número de trabalhos na literatura em relação aos problemas de variação de expressão/iluminação/pose, o problema de oclusão é relativamente negligenciado pela comunidade científica. Embora tenha sido dada pouca atenção ao problema de oclusão na literatura de reconhecimento facial, a importância deste problema deve ser enfatizada, pois a presença de oclusão é muito comum em cenários não controlados e pode estar associada a várias questões de segurança. Por outro lado, a Multibiométria é uma abordagem relativamente nova para representação de conhecimento biométrico que visa consolida múltiplas fontes de informação visando melhorar a performance do sistema biométrico. Multibiométria é baseada no conceito de que informações obtidas a partir de diferentes modalidades ou da mesma modalidade capturada de diversas formas se complementam. Consequentemente, uma combinação adequada dessas informações pode ser mais útil que o uso de informações obtidas a partir de qualquer uma das modalidades individualmente. A fim de melhorar a performance dos sistemas biométricos faciais na presença de oclusão parciais será investigado o emprego de diferentes técnicas de reconstrução de oclusões parciais de forma a gerar diferentes imagens de face, as quais serão combinadas no nível de extração de característica e utilizadas como entrada para um classificador neural. Os resultados demonstram que a abordagem proposta é capaz de melhorar a performance dos sistemas biométricos baseados em face parcialmente ocluídas
With the advancement of technology, traditional strategies for identifying people have become more susceptible to failures. In order to overcome these difficulties, some approaches have been proposed in the literature. Among these approaches, Biometrics stands out. The field of biometrics covers a wide range of technologies used to identify or verify a person\'s identity by measuring and analyzing physical and / or behavioral aspects of the human being. As a result, a biometry has a wide field of applications in systems that require a secure identification of its users. The most popular biometric systems are based on facial recognition or fingerprints. However, there are biometric systems that use the iris, retinal scan, voice, hand geometry, and facial thermograms. Currently, there has been significant progress in automatic face recognition under controlled conditions. In real world applications, facial recognition suffers from a number of problems in uncontrolled scenarios. These problems are mainly due to different facial variations that can greatly change the appearance of the face, including variations in expression, illumination, posture, as well as partial occlusions. Compared with the large number of papers in the literature regarding problems of expression / illumination / pose variation, the occlusion problem is relatively neglected by the research community. Although attention has been paid to the occlusion problem in the facial recognition literature, the importance of this problem should be emphasized, since the presence of occlusion is very common in uncontrolled scenarios and may be associated with several safety issues. On the other hand, multibiometry is a relatively new approach to biometric knowledge representation that aims to consolidate multiple sources of information to improve the performance of the biometric system. Multibiometry is based on the concept that information obtained from different modalities or from the same modalities captured in different ways complement each other. Accordingly, a suitable combination of such information may be more useful than the use of information obtained from any of the individuals modalities. In order to improve the performance of facial biometric systems in the presence of partial occlusion, the use of different partial occlusion reconstruction techniques was investigated in order to generate different face images, which were combined at the feature extraction level and used as input for a neural classifier. The results demonstrate that the proposed approach is capable of improving the performance of biometric systems based on partially occluded faces

Identity verification is one of the most common and important processes in our daily lives. For centuries, humans have relied on the visual appearance of their peers - or other distinctive traits - to recognize them. Currently, most of the traditional authentication methods infer one s identity by either verifying the knowledge of a shared secret (i.e., password) or the possession of a given object (i.e., token); both methods are sub-optimal, since they do not directly verify the person s identity, but an alternate and less rich representation that could also very easily stolen or inadvertently shared. Biometrics offers a natural solution to this problem, by providing quantitative methods for recognizing one s identity by the analysis of either physiological or behavioural traits. In addition to the traditional biometric traits, such as fingerprint, iris or voice, there is a growing interest in novel biometric traits, that can be used in conjunction with the most established ones to compensate to their weaknesses. In the first part of this thesis, we will analyze the usage of heart sounds as a physiological trait for biometric recognition, discussing some novel ad-hoc algorithms developed to process them. Forensic analysts often have to determine whether a given speech sample was uttered by a suspect or not; in the second part of the thesis, we will investigate the usage of automatic text-independent speaker recognition systems in this context, exploring the limits of this approach and proposing new solutions. Finally, given the capillary diffusion that Internet access has gained in the last years, we will analyze the problem of biometric authentication for web application. Through the performance analysis of 3 biometric systems and their combination using 2 multi-biometric score level fusion strategies, we will find the optimal combination of those system; we will then present the architecture and implementation of an open-source web-based multi-biometric authentication system based on speech and face recognition, fused together using the optimal strategy identified in the preliminary analysis phase.

Ingeniero Civil Eléctrico
La biometría corresponde al estudio de medidas en base a diferentes características humanas, tales como huellas digitales, iris, rostro y voz, entre otros. Un gran campo de aplicación de esta corresponde al reconocimiento de rostro para seguridad y control de identidad. Recientemente, se han realizado estudios que indican que la zona peri-ocular, segmento que rodea al ojo, puede ser usado en un sistema de reconocimiento con buenos resultados. Este trabajo de título propone como objetivo estudiar la fusión de información entre dos sistemas de reconocimiento, basado en imágenes de rostro e imágenes de zona peri-ocular, a nivel de características, puntaje y decisión. Para ello se usan las bases de datos AT&T de rostro y una base propia de imágenes de zona peri-ocular pertenecientes al laboratorio de procesamiento de imágenes del Departamento de Ingeniería Eléctrica de la Universidad de Chile. Se implementan sistemas de reconocimiento basándose en 3 métodos de extracción de características diferentes, PCA, LDA y LBP, en C++ utilizando la librería FaceRecognizer de OpenCV. Se implementa además un sistema de fusión para cada nivel de información: característica, puntaje y decisión. Se realizan pruebas de desempeño a los sistemas de reconocimiento de forma individual, fusionados por nivel e integrados totalmente y se comparan mediante el área bajo la curva ROC y la curva de Precision-Recall. Se crea además un sistema fusionado por puntaje válido y moda de decisión llegando a un 98.57% de clasificación correcta. Finalmente se concluye que el mejor tipo de fusión es en el nivel de decisión, considerando el costo computacional de los algoritmos, y se entregan detalles sobre las direcciones a seguir en una posible investigación futura.

This dissertation focuses on person identification problems and proposes solutions to overcome those problems. First part is about fingerprint features extraction algorithm performance evaluation. Modifications to a known synthesis algorithm are proposed to make it fast and suitable for performance evaluation. Matching of deformed fingerprints is discussed in the second part of the work. New fingerprint matching algorithm that uses local structures and does not perform fingerprint alignment is proposed to match deformed fingerprints. The use of group delay features of linear prediction model for speaker recognition is proposed in the third part of the work. New similarity metric that uses group delay features is described. It is demonstrated that automatic speaker recognition system with proposed features and similarity metric outperforms traditional speaker identification systems . Multibiometrics using fingerprints and voice is addressed in the last part of the dissertation.
Penkiose disertacijos darbo dalyse nagrinėjamos žmogaus identifikavimo pagal pirštų atspaudus ir balsą problemos ir siūlomi jų sprendimai. Pirštų atspaudų požymių išskyrimo algoritmų kokybės įvertinimo problemą siūloma spręsti panaudojant sintezuotus pirštų atspaudus. Darbe siūlomos žinomo pirštų atpaudų sintezės algoritmo modifikacijos, kurios leidžia sukurti piršto atspaudo vaizdą su iš anksto nustatytomis charakteristikomis ir požymiais bei pagreitina sintezės procesą. Pirštų atspaudų požymių palyginimo problemos yra aptartos ir naujas palyginimo algoritmas yra siūlomas deformuotų pirštų palyginimui. Algoritmo kokybė yra įvertinta ant viešai prieinamų ir vidinių duomenų bazių. Naujas asmens identifikavimo pagal balsą metodas remiantis tiesinės prognozės modelio grupinės delsos požymiais ir tų požymių palyginimo metrika kokybės prasme lenkia tradicinius asmens identifikavimo pagal balsą metodus. Pirštų ir balso įrašų nepriklausomumas yra irodytas ir asmens atpažinimas pagal balsą ir pirštų atspaudus kartu yra pasiūlytas siekiant išspręsti bendras biometrinių sistemų problemas.

Orientador: Aparecido Nilceu Marana
Banca: Roberto Marcondes Cesar Junior
Banca: Ivan Rizzo Guilherme
Resumo: O reconhecimento facial é uma das principais formas de identificação humana. Apesar das pesquisas em reconhecimento facial automático terem crescido substancialmente ao longo dos últimos 35 anos, identificar pessoas a partir da face continua sendo um desafio para as áreas de Visão Computacional e Reconhecimento de Padrões. Em função dos cenários variarem desde a identificação a partir de fotografias até o reconhecimento baseado em vídeos sem nenhum tipo de controle ao serem gravados, os maiores desafios estão relacionados à independência contra diferentes tipos de iluminação, pose e expressão. O objetivo desta dissertação é propor técnicas que possam contribuir para a melhoria dos sistemas de reconhecimento facial. A primeira técnica endereça o problema da iluminação através da fusão dos espectros visível e infravermelho da face. Através desta abordagem, as taxas de reconhecimento foram melhoradas em 2.07% enquanto a taxa de erro igual (EER) foi reduzida em 45.47%. A segunda técnica trata do caso da extração e classificação de características faciais. Ela propõe um novo modelo para reconhecimento facial através do uso de características extraídas por Histogramas Census e de uma técnica de reconhecimento de padrões baseada em Máquinas de Vetores de Suporte (SVMs). Este outro grupo de experimentos nos possibilitou aumentar a precisão do reconhecimento no teste FERET fa/fb em 0.5%. Além destes resultados, algumas contribuições adicionais deste trabalho que merecem ser destacadas são a análise da dependência estatística entre classificadores de espectros diferentes e considerações sobre o comportamento de uma única C-SVC SVM para identificação de pessoas de forma eficaz.
Abstract: Face recognition is one of the primary ways of human identification. Although researches on automated face recognition have broadly increased along the last 35 years, it remains a challenging task in the fields of Computer Vision and Pattern Recognition. As the scenarios varies from static and constrained photographs to uncontrolled video images, the challenging issues on automatic face recognition are usually related with variations in illumination, pose and expressions. The goal of this master thesis is to propose techniques for the improvement of face recognition systems. The first technique addresses the problem of illumination by fusing the visible and the infrared spectra of the face. With this approach the recognition rates were improved in 2.07% while the Equal Error Rate (EER) were reduced in 45.47%. The second technique addresses the issue of face features extraction and classification. It proposes a new framework for face recognition by using features extracted by Census Histograms and a pattern recognition technique based on Support Vector Machines (SVMs). This other group of experiments enabled us to increase the recognition accuracy in the FERET fa/fb test in 0.5%. Beyond these results, additional contributions of this work that deserve to be highlighted are the statistical dependency analysis between face recognition systems based on different spectra and a better comprehension about the behavior of a single C-SVC SVM to reliably predict faces identities.
Mestre

Orientador: Aparecido Nilceu Marana
Banca: Fátima de Lourdes dos Santos Nunes Marques
Banca: Marcos Antônio Cavenaghi
Resumo: Biometria refere-se ao uso de características físicas (impressões digitais, íris, retina) ou comportamentais (assinatura, voz) para a identificação humana. As impressões digitais são formadas por cristas e minúcias. As cristas são linhas distribuídas paralelamente com uma orientação e um espaçamento característico e as minúcias representam os vários modos pelos quais uma crista pode se tornar descontínua. Graças a sua universalidade, unicidade e permanência, as impressões digitais tornaram-se as características biométricas mais amplamente utilizadas. Entretanto, considerar o reconhecimento automático de impressões digitais um problema totalmente resolvido é um erro muito comum. Nenhum sistema de reconhecimento de impressões digitais proposto até hoje é infalível, nenhum garante taxas de erro nulas. Imagens de baixa qualidade e com pequena área de sobreposição entre a imagem template e a imagem de consulta ainda representam um desafio para os métodos de reconhecimento de impressões digitais mais utilizados, os métodos baseados no casamento de pontos de minúcias. Uma das maneiras de superar as limitações e melhorar a acurácia de um sistema biométrico é o uso da multibiometria, isto é, a combinação de diferentes tipos de informação em um sistema de reconhecimento biométrico. Neste contexto, esta dissertação de mestrado objetiva aprimorar a acurácia dos sistemas de reconhecimento de impressões digitais por meio da fusão de métodos baseados em minúcias e em cristas. Para tanto, foram implementadas técnicas de fusão no nível de pontuação, classificação e decisão. No nível de pontuação, a fusão propiciou uma redução na taxa de erro igual (EER) de 42,53% em relação ao método mais preciso. Para o nível de classificação, a fusão significou um aumento de 75% na taxa de recuperação correta... (Resumo completo, clicar acesso eletrônico abaixo)
Abstract: Biometrics refers to the use of physical (fingerprints, iris, retina) or behavioral (signature, voice) characteristics to determine the identity of a person. Fingerprints are formed by ridges and minutiae. The ridges are lines distributed in parallel with an orientation and a characteristic spacing and the minutiae represent the several ways a ridge can become discontinued. As to its universality, uniqueness and permanence, the fingerprints became the most widely used biometric characteristic. However, it is a common mistake to consider the automatic fingerprint recognition as a totally solved problem. No fingerprint recognition system proposed until now is infallible, none of them guarantee null error rates. Poor quality images and when just a small area of overlap between the template and the query images exists are still a complex challenge to the most used fingerprint recognition methods, the methods based on minutiae points matching. One of the possibilities to overcome the limitations and improve the accuracy of a biometric system is the use of multibiometrics, the combination of different kinds of information in a biometric system. In this context, this master thesis aims to improve the accuracy of fingerprint recognition systems through the fusion of minutiae based and ridge based methods. To achieve this, fusion techniques on score, rank and decision levels were implemented. For the score level, the fusion lead to a reduction of the Equal Error Rate to 42.53% compared to the most precise method. For the rank level, the fusion meant an increase of 75% in the Correct Retrieval Rate. And, in the decision level fusion the Recognition Rate changed from 99.25% to 99.75%. The results have demonstrated that the fusion of minutiae based and ridge based methods can represent a significant accuracy improvement for the fingerprint recognition systems.
Mestre

Made available in DSpace on 2015-05-14T12:36:30Z (GMT). No. of bitstreams: 1 arquivototal.pdf: 4594295 bytes, checksum: bd4f4df655903b796eb6cf79a5060ded (MD5) Previous issue date: 2011-09-16
Coordenação de Aperfeiçoamento de Pessoal de Nível Superior
A biometric system that employs one single biometric characteristic is constrained. This limitation can be reduced by fusing the information presented by multiple sources. A system that consolidates the evidence presented by multiple biometric sources is known as a multibiometric system. In such a context, this work proposes the security and persistence APIs of a multi-biometric architecture, which is capable of using one or more biometric modalities. In access control applications, a user might be forced to authenticate in order to give an unauthorized access to a criminal. As an alternative to this problem, the API uses a continuous authentication process, which verifies if the user identified at the start of the software application is still able to remain on the system, without human interferences or breaks in the process. Much of the literature on biometric system design has focused on system error rates and scaling equations. However, it is also important to have a solid foundation for future progress as the processes and systems architecture for the new biometric application are designed. Hence, the designed architecture made it possible to create a well-defined API for multibiometric systems, which may help developers to standardize, among other things, their data structure, in order to enable and facilitate templates fusion and interoperability. Therefore, the developed security and persistence APIs support a multi-biometric access control architecture. This architecture is extensible, that is, capable of easily comprising new biometric characteristics and processes, yet making it possible to use a template security mechanism. The APIs were designed and implemented. They were demonstrated by a prototype application, through which it was possible to conduct the test experiments.
Um sistema biométrico que empregue uma única peculiaridade ou traço característico é restrito. Esta limitação pode ser suavizada pela fusão dos dados apresentados por múltiplas fontes. Um sistema que consolida a evidência apresentada por múltiplas fontes biométricas é conhecido como um sistema multibiométrico. Nesse contexto, este trabalho propõe a interface de aplicação (API) de segurança e armazenamento de uma arquitetura multibiométrica, com habilidade de empregar uma ou mais modalidades biométricas. Em aplicações de controle de acesso, um usuário pode ser coagido a se autenticar para permitir um acesso indevido. Como alternativa para este problema, a API utiliza um processo de autenticação contínua, que verifica se o usuário que se identificou no início de uma aplicação de software ainda está apto a continuar no sistema, sem interferências humanas ou paralisações do processo. Grande parte da literatura sobre projeto de sistemas biométricos tem o foco nas taxas de erro do sistema e na simplificação de equações. No entanto, também é importante que se tenha uma base sólida para progressos futuros no momento em que os processos e a arquitetura da nova aplicação biométrica estiverem sendo projetados. Neste sentido, a arquitetura projetada permitiu a construção de uma API bem definida para sistemas multibiométricos, que deverá auxiliar os desenvolvedores a padronizar, entre outras coisas, sua estrutura de dados, de forma a possibilitar e facilitar a fusão de modelos biométricos e a interoperabilidade. Deste modo, a API de segurança e armazenamento desenvolvida suporta uma arquitetura multibiométrica de controle de acesso para autenticação contínua extensível, isto é, capaz de receber novas características e processos biométricos com facilidade, permitindo, ainda, o uso de um mecanismo de segurança de templates biométricos. A API foi projetada e implementada. Sua demonstração foi feita através de uma aplicação protótipo, por meio da qual foi possível realizar os testes.

Although the market for biometric technologies is expanding, the existing biometric systems present still some issues that the research community has to address. In particular, in adverse environmental conditions (e.g., low quality biometric signals), where the error rates increase, it is necessary to create more robust and dependable systems. In the literature on biometrics, the integration of multiple biometric sources has been successfully used to improve the recognition accuracy of the unimodal biometric systems. Multibiometric systems, by exploiting more information, such as different biometric traits, multiple samples, multiple algorithms, make more reliable the biometric authentication. Benefits of multibiometrics depend on the diversity among the component matchers and also, on the competence of each one of them. In non-controlled conditions of data acquisition, there is a degradation of biometric signal quality that often causes a significant deterioration of recognition performance. It is intuitive the concept that, the classifier having the higher quality is more credible than a classifier operating on noisy data. Then, researchers started to propose quality-based fusion schemes, where the quality measures of the samples have been incorporated in the fusion to improve performance. Another promising direction in multibiometrics is to estimate the decision reliability of the component modality matcher based on the matcher output itself. An interesting open research issue concerns how to estimate the decision reliability and how to exploit this information in a fusion scheme. From a security perspective, a multimodal system appears more protected than its unimodal components, since spoofing two or more modalities is harder than spoofing only one. However, since a multimodal system involves different biometric traits, it offers a higher number of vulnerable points that may be attacked by a hacker who may choice to fake only a subset of them. Recently, researchers investigated if a multimodal system can be deceived by spoofing only a subset but not all the fused modalities. The goal of this thesis is to improve the performance of the existing integration mechanisms in presence of degraded data and their security in presence of spoof attacks. Our contribution concerns three important issues: 1) Reducing verification errors of a fusion scheme at score level based on the statistical Likelihood Ratio test, by adopting a sequential test and, when the number of training samples is limited, a voting strategy. 2) Addressing the problem of identification errors, by setting up a predictor of errors. The proposed predictor exploits ranks and scores generated by the identification operation and can be effectively applied in a multimodal scenario. 3) Improving the security of the existing multibiometric systems against spoof attacks which involve some but not all the fused modalities. Firstly, we showed that in such a real scenario performance of the system dramatically decrease. Then, for the fingerprint modality, we proposed a novel liveness detection algorithm which combines perspirationand morphology-based static features. Finally, we demonstrated that, by incorporating our algorithm in the fusion scheme, the multimodal system results robust in presence of spoof attacks.