Cette thèse porte sur les techniques d'analyse formelle de systèmes hybrides à dynamiques continues non linéaire. Ses contributions portent sur les algorithmes d'atteignabilité et sur les problèmatiques liées à la representation des ensembles atteignables. This thesis deals with formal analysis of hybrid system with non linear continous dynamic. It contributes to the fields of reachability analysis algorithm and the set representation. / In this thesis, we presented our contributions to the formal analysis of dynamical systems. We focused on the problem of efficiently computing an accurate approximation of the reachable sets under nonlinear dynamics given by differential equations. Our aim was also to design scalable methods which can handle large systems. The first contribution of this thesis concerns the dynamic hybridization technique for a large class of nonlinear systems. We focused on the hybridization domain construction such that the linear interpolation realized in this domain ensures a desired error between the original system trajectories and those computed with the approximated system. We propose a construction method which tends to maximize the domain volume which reduce the number of creation of new domains during the analysis. The second research direction that we followed concerns a subclass of nonlinear dynamical systems which are the polynomial systems. Our results for the reachability analysis of these systems are based on the Bernstein expansion properties. We approximate an initial reachability computation (which requires solving polynomial optimization problems) with an accurate over-approximation (which requires solving linear optimization problems). The last theoretical contribution concerns the reachability analysis of linear systems with polyhedral input which often result from approximation of nonlinear systems. We proposed a technique to refine

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 However, biometric features may change over time, an issue that can affect the recognition performance due to an outdated biometric reference.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.

Made available in DSpace on 2016-08-17T14:53:12Z (GMT). No. of bitstreams: 1 Irlandino Almeida.pdf: 795180 bytes, checksum: 9f8896e26b58665bad7d75b665940404 (MD5) Previous issue date: 2007-09-17 / We present a semi-automatic methodology for three-dimensional reconstruction using an image-based modeling environment. This is done through an initial camera calibration step that makes possible to the viewer, identify objects in the 2D acquired image and them get its position from the camera parameters. This methodology can be applied in the real word model reconstruction like electric and industry installations. / Apresentamos uma metodologia semi-automática para reconstrução tridimensional utilizando um ambiente de modelagem no qual o usuário, a partir de uma determinada imagem e de pontos nela selecionados com seu correspondente no espaço do mundo, realiza a calibração da câmera. Após este processo, a cena poderá ser composta a partir de objetos pré-estabelecidos fornecidos pelo ambiente de modelagem tridimensional baseado na imagem. Esta metodologia pode ser aplicada na reconstrução de modelos do mundo real, como instalações elétricas e industriais.

Submitted by Renata Lopes (renatasil82@gmail.com) on 2017-06-07T14:26:02Z No. of bitstreams: 1 helenadealmeidamaia.pdf: 3132814 bytes, checksum: d46a470b453ec6ba11362abaeac3a42c (MD5) / Approved for entry into archive by Adriana Oliveira (adriana.oliveira@ufjf.edu.br) on 2017-06-07T14:56:50Z (GMT) No. of bitstreams: 1 helenadealmeidamaia.pdf: 3132814 bytes, checksum: d46a470b453ec6ba11362abaeac3a42c (MD5) / Made available in DSpace on 2017-06-07T14:56:50Z (GMT). No. of bitstreams: 1 helenadealmeidamaia.pdf: 3132814 bytes, checksum: d46a470b453ec6ba11362abaeac3a42c (MD5) Previous issue date: 2016-03-18 / CAPES - Coordenação de Aperfeiçoamento de Pessoal de Nível Superior / Nos últimos anos, o rastreador TLD (Tracking-Learning-Detection) se destacou por combinar um método de rastreamento através do movimento aparente e um método de detecção para o problema de rastreamento de objetos em vídeos. O detector identifica o objeto pelas aparências supostamente confirmadas. O rastreador insere novas aparências no modelo do detector estimando o movimento aparente. A integração das duas respostas é realizada através da mesma métrica de similaridade utilizada pelo detector que pode levar a uma decisão enviesada. Neste trabalho, é proposto um framework para métodos baseados em múltiplos rastreadores onde o componente responsável pela integração das respostas é independente dos rastreadores. Este componente é denominado mediador. Seguindo este framework, um novo método é proposto para integrar o rastreador por movimento e o detector do rastreador TLD pela combinação das suas estimativas. Os resultados mostram que, quando a integração é independente das métricas de ambos os rastreadores, a performance é melhorada para objetos com significativas variações de aparência durante o vídeo. / On the problem of tracking objects in videos, a recent and distinguished approach combining tracking and detection methods is the TLD (Tracking-Learning-Detection) framework. The detector identifies the object by its supposedly confirmed appearances. The tracker inserts new appearances into the model using apparent motion. Their outcomes are integrated by using the same similarity metric of the detector which, in our point of view, leads to biased results. In our work, we propose a framework for generic multitracker methods where the component responsible for the integration is independent from the trackers. We call this component as mediator. Using this framework, we propose a new method for integrating the motion tracker and detector from TLD by combining their estimations. Our results show that when the integration is independent of both tracker/detector metrics, the overall tracking is improved for objects with high appearance variations throughout the video.

Com a popularização de equipamentos tradicionais de captura de imagens, como câmeras digitais, e o avanço tecnológico dos dispositivos não invasivos, como tomografia e ressonância, cresce também a necessidade e consequente uso de métodos numéricos para simulação de fenômenos físicos em domínios definidos por imagens. Um dos pré-requisitos para a aplicação de tais métodos numéricos consiste na discretização do domínio em questão, num processo denominado geração de malhas. Embora diversos métodos de geração de malha tenham sido propostos para discretizar domínios definidos por primitivas geométricas, pouco tem sido feito no sentido de gerar uma decomposição diretamente a partir de imagens. Neste trabalho, apresentamos uma abordagem de geração de malhas de qualidade a partir de domínios definidos por imagens com textura. Mais especificamente, a pesquisa descrita nesta tese contribui com a melhoria do algoritmo Imesh, ao sanar três de suas principais limitações: tratamento de imagens com texturas; controle do nível de refinamento da malha e suporte a outros tipos de elementos. Estas contribuições flexibilizam o processo de geração da malha, e ainda ampliam o domínio de aplicações do algoritmo Imesh, à medida que são considerados domínios definidos por imagens com textura e o uso de métodos numéricos para elementos não simpliciais torna-se possível. O algoritmo de melhoria da malha gerada utiliza uma abordagem inovadora de remalhamento baseada em templates e guiada por retalhos de Bézier / With the spreading of traditional image capturing devices, such as digital cameras, and the technological advancement of more specific imaging devices such as CT and MRI, also increased the need and the following use of numerical methods for simulation of physical phenomena in domains defined by images. One of the prerequisites for the application of such numerical methods is the discretization of the corresponding domain, in a process called mesh generation. Although several methods of mesh generation have been proposed to discretize domains defined by geometric primitives, little has been done to generate a decomposition directly from images. We present an approach to generate quality meshes from domains defined by images with texture. More specifically, the research described in this thesis contributes to the improvement of the Imesh algorithm, removing three of its main limitations: treatment textured images, control of the level of the mesh refinement and support for other types of non-simplicial elements. These contributions provide flexibility to the mesh generation process, and extend the range of applications of Imesh by both handling textured images and considering the use of numerical methods for non-simplicial elements. The mesh quality improvement algorithm uses a new approach based on mesh templates and it is guided by Bezier patches

Este trabalho descreve a Klein, uma biblioteca genérica para álgebra linear em C++. A Klein facilita o uso de matrizes e vetores, permitindo que o usuário programe de modo similar ao Matlab. Com ela podemos, por exemplo, implementar um passo do método de Newton para a função f, através da expressão x = x - inv(jac(x)) * f(x), onde x é o vetor, jac a Jacobiana e inv a inversa. Além disso, por se tratar de uma biblioteca genérica, os tipos envolvidos nestas expressões podem ser escolhidos pelo programador. O trabalho também discute como a biblioteca é testada, tanto do ponto de vista de corretude quanto de desempenho. / We describe the Klein library, a generic libray for linear algebra in C++. It simplifies the use of vectors and matrices and let the user program as in Matlab. With Klein, one can for instance implement Newton\'s method as x = x - inv(jac(x)) * f(x), where x is a vector, jac is the Jacobian matrix, inv is the inverse operator and f(x) is the function of which we want to find zero. Moreover, Klein is generic in the sense that it allows the use of arbitrary types of scalars (float, double, intervals, rationals, etc). We also explain how it is tested, both for correctness and performance.

In the scheme of Non Destructive Testing (NDT), defect detection is an important process. Traditional image processing techniques have successfully been used for defect recognition. Usage of machine learning techniques is still in the initial stages of development. Convolution Neural Networks (CNN) is widely used for object classification one such scenario is defect classification in weld tubes. With the advent of deep learning techniques such as transfer learning, we can transfer knowledge gained in one domain successfully into other. Pre-trained models successfully learn features from large scale datasets that can be used for in domains having sparse data and smaller datasets. The aim of this work is to help a manual inspector in recognition of defects on the weld tubes. With a given set of images, we proceed by forming unique pipeline architecture for automatic defect recognition. The research in this thesis focuses on extraction of welds using image segmentation techniques, creating a dataset of defects and using it to on pre-trained Convolution Neural Networks of VGG16, VGG19, Inception V3 and ResNet101. We evaluate the models on different metrics finding the best suited model for the created dataset. Further a prototype sliding window solution is used to find defects over the extracted weld region. We also present the limitations of this approach and suggest modifications that can be implemented in the future.

Abstract: When noble metals particles are synthesized with progressively smaller dimensions, strikingly novel optical properties arise. For nanoscale particles, collective disturbances of the electron density known as localized surface plasmons resonances can arise, and these resonances are utilized in a variety of applications ranging from surface-enhanced molecular spectroscopy and sensing to photothermal cancer therapy to plasmon-driven photochemistry. Central to all of these studies is the plasmon’s remarkable ability to process light, capturing and converting it into intense near fields, heat, and even energetic carriers at the nanoscale. In the past decade, we have witnessed major advances in plasmonics which is directly linked with the much broader field of (colloidal) nanotechnology. These breakthroughs span from plasmon lasing and waveguides, plasmonic photochemistry and solar cells to active plasmonics, plasmonics nanocomposites and semiconductor plasmons. All the above-mentioned phenomena rely on precise spatial placement and distinct control over the dimensions and orientation of the individual plasmonic building blocks within complex one-, two- or three-dimensional complex arrangements. For the nanofabrication of metal nanostructures at surfaces, most often lithographic approaches, e.g. e-beam lithography or ion-beam milling are generally applied, due to their versatility and precision. However, these techniques come along with several drawbacks such as limited scalability, limited resolution, limited compatibility with silicon manufacturing techniques, damping effects due to the polycrystalline nature of the metal nanostructures and low sample throughput. Thus, there is a great demand for alternative approaches for the fabrication of metal nanostructures to overcome the above-mentioned limitations. But why colloids? True three-dimensionality, lower damping, high quality modes due to mono-dispersity, and the absence of grain boundaries make the colloidal assembly an especially competitive method for high quality large-scale fabrication. On top of that, colloids provide a versatile platform in terms of size, shape, composition and surface modification and dispersion media. 540The combination of directed self-assembly and laser interference lithography is a versatile admixture of bottom-up and top-down approaches that represents a compelling alternative to commonly used nanofabrication methods. The objective of this thesis is to focus on large area fabrication of emergent spectroscopic properties with high structural and optical quality via colloidal self-assembly. We focus on synergy between optical and plasmonic effects such as: (i) coupling between localized surface plasmon resonance and Bragg diffraction leading to surface lattice resonance; (ii) strong light matter interaction between guided mode resonance and collective plasmonic chain modes leading to hybrid guided plasmon modes, which can further be used to boost the hot-electron efficiency in a semiconducting material; (iii) similarly, bilayer nanoparticle chains leading to chiro-optical effects. Following this scope, this thesis introduces a real-time tuning of such exclusive plasmonic-photonic (hybrid) modes via flexible template fabrication. Mechanical stimuli such as tensile strain facilitate the dynamic tuning of surface lattice resonance and chiro-optical effects respectively. This expands the scope to curb the rigidity in optical systems and ease the integration of such systems with flexible electronics or circuits.:Contents Abstract Kurzfassung Abbreviations 1. Introduction and scope of the thesis 1.1. Introduction 1.1.1. Classical optics concepts 1.1.2. Top down fabrication methods and their challenges 1.1.3. Template-assisted self-assembly 1.1.4. Functional colloidal surface assemblies 1.2. Scope of the thesis 2. Results and Discussion 2.1. Mechanotunable Surface Lattice Resonances in the Visible Optical Range by Soft Lithography Templates and Directed Self-Assembly 2.1.1. Fabrication of flexible 2D plasmonic lattice 2.1.2. Investigation of the influence of particle size distribution on SLR quality 2.1.3. Band diagram analysis of 2D plasmonic lattice 2.1.4. Strain induced tuning of SLR 2.1.5. SEM and force transfer analysis in 2D plasmonic lattice under various strain 2.2. Hybridized Guided-Mode Resonances via Colloidal Plasmonic Self-Assembled Grating 2.2.1. Fabrication of hybrid opto-plasmonic structure via template assisted self-assembly 2.2.2. Comparison of optical band diagram of three (plasmonic, photonic and hybrid) different structures in TE and TM modes 2.2.3. Simulative comparison of optical properties of hybrid opto-plasmonic NP chains with a grating of metallic gold bars 2.2.4. Effect of cover index variation with water as a cover medium 2.3. Hot electron generation via guided hybrid modes 2.3.1. Fabrication of the hybrid GMR structure via LIL and lift-off process 2.3.2. Spectroscopic and simulative analysis of hybrid opto-plasmonic structures of different periodicities 2.3.3. Comparative study of photocurrent generation in different plasmonic structures 2.3.4. Polarization dependent response at higher wavelength 2.3.5. Directed self-assembly of gold nanoparticles within grating channels of a dielectric GMR structure supported by titanium dioxide film 2.4. Active Chiral Plasmonics Based on Geometrical Reconfiguration 2.4.1. Chiral 3D assemblies by macroscopic stacking of achiral chain substrates 3. Conclusion 4. Zusammenfassung 5. Bibliography 6. Appendix 6.1. laser interference lithography 6.2. Soft molding 6.3. Determine fill factor of plasmonic lattice 6.4. 2D plasmonic lattice of Au_BSA under strain 6.5. Characterizing order inside a 2D lattice 6.6. Template-assisted colloidal self-assembly 6.7. Out of plane lattice resonance in 1D and 2D lattices 6.8. E-Field distribution at out of plane SLR mode for 1D lattices of various periodicity with AOI 20° 6.9. Refractive index of PDMS and UV-PDMS 6.10. Refractive index measurement for sensing 6.11. Optical constants of TiO2, ma-N 405 photoresist and glass substrate measured from spectroscopic ellipsometry Acknowledgement/ Danksagung Erklärung & Versicherung List of Publications

This thesis is focused on face detection in static picture. Theoretical part contains color spaces (RGB, HSI, YCbCr), methods for skin detection (explicit, parametric or non-parametric methods), image metric, edge detection, mathematical morphology, methods for classification faces (appearance-based methods, feature invariant approaches, knowledge-based methods, template matching methods). Practical part of this thesis contains concept and practical realization two algorithms for segmentation skin in static image (simple method based on Cr chroma components and statistical method). Practical part contains concept and practical realization two algorithms for classification face (appearance-based method and template matching method) too.

Based on the increasing complexity of modern avionics, the associated system design processes moved towards Model Driven Architecture (MDA) based processes. Additionally, the demand for higher system autonomy features requires means to further modularise mission systems and to define and establish interactions among the systems’ individual components. Therefore, the ideas of service-oriented computing are currently adapted to established, model driven design processes. With respect to modelling interactions for service components, current approaches are limited to only a fixed set of communication primitives, restricting a service designer’s expressiveness to specify service interaction. In consequence, interaction patterns not included in this basic set have to be reflected in application code, mangling application and communication logic. Furthermore, when service functionality relies on communication semantics which are not provided by the underlying set of primitives, additional emulation behaviour has to be added to the service which makes this mangling even worse. Platform independence is reduced as services can not easily be ported to platforms not natively supporting the selected primitives which contradicts the ideas of model-driven development. Addressing these limitations, this thesis proposes a new model-driven service development process based on Interaction Templates (ITs), promoting interactions among service participants to first class modelling entities. The process focuses on modelling the interactions among service participants. Interaction semantics are explicitly specified in models, beyond pure stereotyping, and gain increased platform independence for services with respect to communication. The process exploits automated Model-to-Model (M2M) and Model-to-Text (M2T) transformations to assist service implementation and to automatically derive interaction realisations on concrete target platforms. This allows for easy replacement and inter-mixing of communication middleware to realise a service’s interactions. This way, services become independent of the underlying communication primitives by only relying on ITs and not platform primitives which are hidden behind ITs. In turn, realising ITs on concrete platforms is not affected by their utilisation for service interaction. Beside the novel modelling process itself, the presented work defines a Unified Modeling Language (UML) profile, referred to as UML Profile for Interaction-centric Services (UP4IS), which directly supports the adaptation of standard UML language constructs and tools for the proposed modelling approach. The whole development process is demonstrated via the specification of a simple video recording systems consisting of two services. The services themselves are based on a representative IT library which forms an essential part of the presented case study. Using these service and IT models, the thesis emphasises the necessary model transformation and code generation steps to derive service implementations based on the abstract models.