Capes / Esta pesquisa estuda a utilização da rede neural Máquina de Aprendizado Extremo (ELM) para identificação de pessoas (biometria) através do eletrocardiograma (ECG). Os dados biométricos oferecem um nível elevado de segurança para a identificação de pessoas, e o ECG é uma técnica emergente e em crescente desenvolvimento. A ELM foi pouco empregada em sistemas de reconhecimento de padrões que utilizam o sinal de ECG. Desta forma, foram estudadas as técnicas de processamento de sinal: a Transformada Wavelet e a Análise dos Componentes Principais (PCA), com o objetivo de tratar e reduzir a dimensionalidade dos dados de entrada, bem como, fazer um estudo comparativo entre a ELM e a Percepetron Múltiplas Camadas (Multilayer Perceptron – MLP). Os testes foram realizados com 90 pessoas, o sinal de ECG utilizado é referente à derivação I contendo 500 amostras/s e 12-bits de resolução dentro de uma faixa nominal de ±10mV de variação, o número de registros variou de 2 a 20 para cada pessoa. O tamanho de cada ciclo completo de ECG para o processo de formação do espaço amostral foi definido de duas formas: 167 amostras contendo as ondas P+QRS e 280 amostras contendo as ondas P+QRS+T, dos quais foram utilizados os 10 ciclos que possuíam o mais elevado nível de similaridade. Com a Transformada Wavelet, o sinal de ECG foi decomposto em 3 níveis, onde para as ondas P+QRS as reduções foram de 86, 45 e 25 amostras, e para as ondas P+QRS+T foram de 142, 73 e 39 amostras. Já para o PCA o sinal foi reduzido de 10 ciclos cardíacos para apenas 1. Estes foram apresentadas a rede formando os conjuntos de treinamento e teste. Foram utilizadas as Redes Neurais Artificiais ELM e MLP para classificação do ECG. Os resultados obtidos comprovaram que a ELM pode ser utilizada para identificação de pessoas. / This research studies the use of neural network Extreme Learning Machine (ELM) to identify individuals (biometrics) by electrocardiogram (ECG). Biometric data offer a high level of security for identifying people, and ECG is an emerging technique and increasing development. ELM was little used in pattern recognition systems that use the ECG signal. In this way, the signal processing techniques were studied: Wavelet Transform and Principal Component Analysis (PCA), with the objective of treating and reducing the dimensionality of the input data, as was as, to make a comparative study between the ELM and Multilayer Perceptron (MLP). The tests were performed with 90 people, the ECG signal used is related to the lead I containing 500 samples/s and 12- bit resolution within a nominal range of ±10 mV of variation, the number of records ranged from 2 to 20 for each people. The size of each ECG cycle to complete the process of forming the sample space defined in two ways: 167 samples containing the P+QRS waves and 280 samples containing the P+QRS+T waves, of which 10 cycles were used to had the highest level of similarity. With the Wavelet Transform, the ECG signal was decomposed into 3 levels, where for the P+QRS waves the reductions were 86, 45 and 25 samples, and for the P+QRS+T waves were 142, 73 and 39 samples. For PCA, the signal for reduced from 10 cardiac cycles to only 1. These were presented to network forming the joint training and testing. The Artificial Neural Networks ELM and MLP were used for ECG classification. The results obtained proved that the ELM may be used to identify individuals.

M.Sc. (Information Technology) / Please refer to full text to view abstract

M.Phil. (Electrical and Electronic Engineering) / Network security has become increasingly important as more and more applica- tions are making their way into the market. The research community has proposed various methods to build a reliable network intrusion detection system to detect unauthorised activities in networked systems. However many network intrusion detection systems that have been reported in literature su er from an excessive number of false positives, false negatives, and are unable to cope with new, elegant and structured attacks. This is mainly because most network intrusion detection systems rely on security experts to analyze the network tra c data and manually construct intrusion detection rules. This study proposes to use a machine learning technique such as neural network approach to anomaly based network intrusion detection system (NIDS). The main objective for this study is to construct an NIDS model that will produce approx- imate to zero false positive or no false positive at all and have high degree of accuracy in detecting network attacks. The neural network (NN) model is trained on a biometric networked system dataset simulated in the study, containing strictly replayed and normal network tra c that encourage the development of the pro- posed NIDS. By analyzing the NN{based NIDS results, the study reached the false positive rate of 0, and high accuracy rate of 100 percent. To support the results obtained in this study, the performance of the NN{based NIDS was compared to two other classi cation methods (k{nearest neighbor algorithm (KNN) and Naive Bayes). The results obtained from KNN and naive Bayes were 99.87 and 99.75 percent respectively. These results show that the proposed model can successfully be used as an e ective tool for solving complicated classi cation problems such as NIDS.

This master thesis evaluates the use of off-the-shelf standard cameras for biometric identification of the human iris. As demands on secure identification are constantly rising and as the human iris provides with a pattern that is excellent for identification, the use of inexpensive equipment could help iris recognition become a new standard in security systems. To test the performance of such a system a review of the current state of the research in the area was done and the most promising methods were chosen for evaluation. A test environment based on open source code was constructed to measure the performance of iris recognition methods, image quality and recognition rate. In this paper the image quality of a database consisting of images from a standard camera is assessed, the most important problem areas identified, and the overall recognition performance measured. Iris recognition methods found in literature are tested on this class of images. These together with newly developed methods show that a system using standard equipment can be constructed. Tests show that the performance of such a system is promising.

We report on the design, implementation and performance issues of the DRUBIS (Distributed Rhodes University Biometric Identification System) experimentation framework. The Principal Component Analysis (PCA) face-recognition approach is used as a case study. DRUBIS is a flexible experimentation framework, distributed over a number of modules that are easily pluggable and swappable, allowing for the easy construction of prototype systems. Web services are the logical means of distributing DRUBIS components and a number of prototype applications have been implemented from this framework. Different popular PCA face-recognition related experiments were used to evaluate our experimentation framework. We extract recognition performance measures from these experiments. In particular, we use the framework for a more indepth study of the suitability of the DFFS (Difference From Face Space) metric as a means for image classification in the area of race and gender determination.

This dissertation proposes the development of an identification verification model that can be implemented within the context of the Eastern Cape, South Africa in order to ensure that vulnerable children are provided with the requisite care that they deserve from the state. The dissertation provides both a developed and developing world perspective on the identification verification needs of vulnerable children by providing an overview of relevant South African policy with regard to caring for vulnerable children and presenting an international perspective with specific reference to current legislative developments in the United Kingdom and Malaysia. Chapter 1 provides a motivation for a framework to be used for the identification verification of children in developing countries by emphasising that the provision of basic social services to children is an urgent requirement for poverty eradication and is a necessity as documented in the United Nations Convention on the Rights of the Child. A background to the needs of vulnerable children in South Africa is given and the scope, limitations and research methodology used in the dissertation is presented. Chapter 2 provides an overview of child related policy in the South African Context both from a National Government and Eastern Cape perspective. Although extensive progress has been made in the development of policies aimed at protecting vulnerable children, the practical implementation of these policies has been hampered by numerous issues including the lack of coordination between key entities. Chapter 3 provides an introduction to several noteworthy international developments with regard to the identity verification of vulnerable children. Lessons learnt from identity verification systems from the United Kingdom and Malaysia are analyzed for applicability to the South African context. In addition to this, the use of biometric technology in identity verification systems and a number of biometric identification methodologies available are discussed. Chapter 4 proposes the development and implementation of a biometric identity verification model in the Eastern Cape Province of South Africa based on lessons learnt from the assessment of South African policy and international best practice. The system should be piloted in the Eastern Cape and, if successful, be implemented throughout South Africa with a possible view to future implementation on the African continent. The scope of the system, the technological requirements and a high level implementation plan together with the need to further research certain key aspects e.g. the cost implications are discussed. It is clear that the development of such a model and the implementation of such a system will ensure that vulnerable children are provided with the requisite care that they are constitutionally entitled to. Significant follow up research is required during the development of the model to ensure that all aspects of the model are well documented and during the implementation of the system to ensure that the requirements of the users both within the government and the general public are met.

M.Sc. (Computer Science) / The past couple of years have seen an explosion in the number of online fraud schemes – Total annual losses exceed tens of millions of Rands. Many people and organizations from all over the world have fallen victim to it. Nobody is safe; everybody is vulnerable. As we increasingly make use of the Internet and mobile technology to do our work and to perform chores such as online banking or shopping, we become even more vulnerable. Fraudsters make use of ever-more sophisticated techniques and clever schemes to target the unsuspecting end-users of mobile and Internet technology, and to trick them into surrendering their personal information so that electronic transactions can be carried out in their name. Examples include cheque and credit card fraud, phishing, identity theft and spyware, to name but a few. Services such as Internet- and cell phone banking especially, provide a haven of possibilities to make easy money because the technology is relatively new, and it is being used by evermore people who believe that it is fast, safe and secure and will therefore make their lives easier. One of the latest scams at the time of this writing is “SIM swap fraud”: Fraudsters target specific victims, perform SIM swaps on their cell phone numbers, and then proceed to empty the victims’ bank accounts. This is all done in a matter of minutes, so the victims only realize what has happened when it is too late to do anything about it. Needless to say, a solution must be found that will prevent unauthorized SIM swaps and strengthen online banking security. This dissertation does exactly that. It examines the digital world known as cyberspace, identifies how SIM swap fraudsters manage to defraud their targets, and presents a biometrics-based security system to combat SIM swap fraud.

Orientadores: Alexandre Xavier Falcão, Giovani Chiachia / Dissertação (mestrado) - Universidade Estadual de Campinas, Instituto de Computação / Made available in DSpace on 2018-08-27T14:41:07Z (GMT). No. of bitstreams: 1 Martins_SamuelBotter_M.pdf: 4782261 bytes, checksum: 63cd58756e3fe70ffe625d42974b1a78 (MD5) Previous issue date: 2015 / Resumo: Sistemas automáticos de reconhecimento de faces tem atraído a atenção da indústria e da academia, devido à gama de possíveis aplicações, tais como vigilância, controle de acesso, etc. O recente progresso em tais sistemas motiva o uso de técnicas de aprendizado em profundidade e classificadores específicos para cada usuário em cenários de operação não-controlado, que apresentam variações consideráveis em pose, iluminação, etc. Sistemas automáticos de reconhecimento de faces possibilitam construir bases de imagens anotadas por meio do processo de cadastramento de novos usuários. Porém, à medida que as bases de dados crescem, torna-se crucial reduzir o número de amostras negativas usadas para treinar classificadores específicos para cada usuário, devido às limitações de processamento e tempo de resposta. Tal processo de aprendizado discriminativo durante o cadastramento de novos indivíduos tem implicações no projeto de sistemas de reconhecimento de faces. Apesar deste processo poder aumentar o desempenho do reconhecimento, ele também pode afetar a velocidade do cadastramento, prejudicando, assim, a experiência do usuário. Neste cenário, é importante selecionar as amostras mais informativas buscando maximizar o desempenho do classificador. Este trabalho resolve tal problema propondo um método de aprendizado discriminativo durante o cadastramento de usuários com o objetivo de não afetar a velocidade e a confiabilidade do processo. Nossa solução combina representações de alta dimensão com um algoritmo que rapidamente minera imagens faciais negativas de um conjunto de minerção grande para assim construir um classificador específico para cada usuário, baseado em máquinas de vetores de suporte. O algoritmo mostrou ser robusto em construir pequenos e eficazes conjuntos de treinamento com as amostras negativas mais informativas para cada indivíduo. Avaliamos nosso método em duas bases contendo imagens de faces obtidas no cenário de operação não-controlado, chamadas PubFig83 e Mobio, e mostramos que nossa abordagem é capaz de alcançar um desempenho superior em tempos interativos, quando comparada com outras cinco abordagens consideradas. Os resultados indicam que o nosso método tem potencial para ser explorado pela indústria com mínimo impacto na experiência do usuário. Além disso, o algoritmo é independente de aplicação, podendo ser uma contribuição relevante para sistemas biométricos que visam manter a robustez à medida que o número de usuários aumenta / Abstract: Automatic face recognition has attracted considerable attention from the industry and academy due to its wide range of applications, such as video surveillance, access control, online transactions, suspect identification, etc. The recent progress in face recognition systems motivates the use of deep learning techniques and user-specific face representation and classification models for unconstrained scenarios, which present considerable variations in pose, face appearance, illumination, etc. Automatic face recognition systems make possible to build annotated face datasets through user enrollment. However, as the face datasets grow, it becomes crucial to reduce the number of negative samples used to train user-specific classifiers, due to processing constraints and responsiveness. Such a discriminative learning process during the enrollment of new individuals has implications in the design of face recognition systems. Even though it might increase recognition performance, it may affect the speed of the enrollment, which in turn may affect the user experience. In this scenario, it is important to select the most informative samples in order to maximize the performance of the classifier. This work addresses this problem by proposing a discriminative learning method during user enrollment with the challenges of not negatively affecting the speed and reliability of the process, and so the user experience. Our solution combines high-dimensional representations from deep learning with an algorithm for rapidly mining negative face images from a large mining set to build an effective classification model based on linear support vector machines for each specific user. The negative mining algorithm has shown to be robust in building small and effective training sets with the most informative negative samples for each given individual. We evaluate our approach on two unconstrained datasets, namely PubFig83 and Mobio, and show that it is able to attain superior performance, within interactive response times, as compared to five other baseline approaches that use the same classification scheme. The results indicate that our approach has potential to be exploited by the industry with minimum impact to the user experience. Moreover, the algorithm is application-independent. Hence, it may be a relevant contribution for biometric systems that aim to maintain robustness as the number of users increases / Mestrado / Ciência da Computação / Mestre em Ciência da Computação

No description available.

The rapid advance of information technology has allowed for a rise in the use of biometric markers to automatically track the identity of individuals. Iris biometrics has emerged as one of the most reliable and accurate systems when dealing with cooperating subjects, however, challenges arise when attempting to minimize the amount of intrusion when examining subjects. Allowing for more flexibility in data capture settings will introduce differences in the iris texture due to changes in ambient light, which may negatively impact recognition results. This research examines the feasibility of using 3D software to synthetically dilate the pupils of existing iris images to more closely match the size of a target image. Methods are developed first to evaluate the compatibility of synthetic images with iris identification software, and then to examine what specific areas of the iris texture differ between synthetic and real images. Results show synthetic images are found to be compatible with the recognition process and have the potential to improve performance.