Em nossa vida diária, são utilizadas identidades digitais (IDDs) para acessar contas de e-mail, bancos e lojas virtuais, locais restritos, computadores compartilhados, e outros. Garantir que apenas usuários autorizados tenham o acesso permitido é um aspecto fundamental no desenvolvimento destas aplicações. Atualmente, os métodos de controle de acesso simples como senhas ou números de identificação pessoal não devem ser considerados suficientemente seguros, já que um impostor pode conseguir estas informações sem o conhecimento do usuário. Ainda, no caso de utilização de dispositivos físicos como cartões de identificação, estes podem ser roubados ou forjados. Para tornar estes sistemas mais confiáveis, técnicas de autenticação de identidades utilizando múltiplas verificações são propostas. A utilização de características biométricas surge como a alternativa mais confiável para tratar este problema, pois são, teoricamente, únicas para cada pessoa. Contudo, algumas características biométricas como a aparência facial podem variar com o tempo, implicando em um grande desafio para os sistemas de reconhecimento facial. Neste trabalho é combinado o acesso tradicional por senha com a análise da face para realizar a autenticação. Um método de aprendizagem supervisionada é apresentado e sua adaptação é baseada na melhora contínua dos modelos faciais, que são representados por misturas de gaussianas. Os resultados experimentais, obtidos sobre um conjunto de teste reduzido, são encorajadores, com 98% de identificação correta dos usuários e custo computacional relativamente baixo. Ainda, a comparação com um método apresentado na literatura indicou vantagens do método proposto quando usado como um pré-selecionador de faces. / In our daily life, we use digital identities (DIDs) to access e-mails, e-banks, e-shops, physical environments, shared computers, and so on. Guarantee that only authorized users are granted access is an important aspect in the development of such applications. Nowadays, the simple access control methods like passwords or personal identification numbers can not be considered secure enough, because an impostor can obtain and use these information without user knowledge. Also, physical devices like ID cards can be stolen. To make these systems more reliable, multimodal DID authentication techniques combining different verification steps are proposed. Biometric features appears as one of the most reliable alternatives to deal with this problem because, theoretically, they are unique for each person. Nevertheless, some biometric features like face appearances may change in time, posing a serious challenge for a face recognition system. In this thesis work, we use the traditional password access combined with human face analysis to perform the authentication task. An intuitive supervised appearance learning method is presented, and its adaptation is based on continuously improving face models represented using the Gaussian mixture modeling approach. The experimental results over a reduced test set show encouraging results, with 98% of the users correctly identified, with a relatively small computational effort. Still, the comparison with a method presented in the literature indicated advantages of the proposed method when used as a pre-selector of faces.

Handwriting as a unique expression of human behaviour has evoked continuous interest as a means of analyzing and studying personality – a study known as Graphology. Research in graphology has shown diverse results, ranging from negative to highly favourable. Many of the studies disregarding the value of graphology can be criticized on the grounds of their research methodology as well as the method used in handwriting analysis, namely the ‘trait-method’ whereby isolated graphological features are simplistically linked to personality as opposed to a more encompassing, holistic approach. In the present study the special link between personality and graphology is illustrated as well as the decisive influence the gestalt and form standard of a writing have on the interpretation of that particular writing. The approach to the graphological analyses was holistic, as it complements the complex uniqueness of personality. The study also focuses on the recommended methodology of handwriting analysis, by offering a tailor-made personality ‘picture’ of the individual. Graphological findings were compared to the clinical findings of the same client/patient according to the Diagnostic and Statistical Manual of Mental Diseases IV-TRTM (DSM-IV-TRTM). Five (5) case studies have been reported in this qualitative research study. The analyses of the writings were done by an independent graphologist. The objective graphological findings show similarities with clinical findings of the same clients/patients. The holistic approach to handwriting analysis deems to have diagnostic value and is promising in providing guidelines for psychotherapy. Graphology can thus be regarded as a useful and viable tool in psycho-educational assessment. / Education / D.Ed. (Psychology of Education)

The cadastral district of Čejkovice was chosen for mapping of landscape features. The district is located approximately 9 km to the north-west of České Budějovice on a surface of 955 hectares. The following materials were used to map the territory in question ? the plan of the Territorial System of Ecological Stability and the orthophotograph map available at geoportal.gov and historical aerial photos from the Military Geographical and Hydrometeorological Institute of Dobruška. The incidence and representation of individual features were ascertained during mapping of the territory and the features were categorized. The categorization serves to process the results ? to evaluate the frequency and surface (length) of the features and at the same time to compare the present status with the status of the 1950?s. The GIS (Geographic Information Systems) software program was used to evaluate the results. At present, 38 landscape features can be found in the territory of Čejkovice (1 watercourse, 9 ponds, 4 woods, 8 woody plant groups, 9 alleys and 7 solitary plants. In 1950?s, 29 features could be found in the territory (1 watercourse, 10 ponds, 3 woods, 6 woody plant groups, 8 alleys and 1 solitary plant). The solitary plants have experienced the greatest difference ? increase by 6 plants. The surface of ponds has been increased as compared to the past status and the surface of woods has grown as well. The condition of the features does not differ too much as compared to the past condition. The most landscape features have experienced growth of surface.

The objective of the submitted study was to find out and assess the specifics of transcultural features, rules and cultural patterns at dying and death in selected minorities in the Czech Republic in relation to nursing care, and to describe the experience of nurses with transcultural nursing and to evaluate the possibilities of using the Transcultural Assessment Model Joyce Newman Giger and Ruth Elaine Davidhizar and the Conceptual Model of Cultural Competencies Larry D. Purnell in transcultural aspects, rules and cultural patterns of dying and death in selected minorities in the Czech Republic. The stated objectives were achieved by a combination of qualitative and quantitative research methods, which were part of the research conducted with the financial support of the Grant Agency of the University of South Bohemia (GAJU) in České Budějovice within the team grant project called "Using Conceptual Models in Clinical and Community Practice ", No 048/2015/S. In the first part of the research, a pre-research with 50 informants was carried out by a semi-structured interview technique, each time with 10 representatives of Judaism, Christianity, Buddhism, Islam and atheism. The main objective of the pre-research was to specify the key data for answering research questions and for creating a self-constructed questionnaire that served together with the standardized DAP-R (The Death Attitudes Profile - Revised) questionnaire to complete quantitative data that analyzed the results of 1 700 respondents. The results of the pre-research study became the basis for the nursing documentation proposal, which was verified in the next phase of the qualitative research in a focus group interview, in which 35 nurses from clinical and community practice were involved. Based on the results of the pre-research study, it was found that the care of the dying and the dead is based on traditional values of human life. Informants demand that nurses have 7 virtues in the form of humility, generosity, compassion, peacefulness, chastity, temperance and activity, enriched with respect for human dignity, love, empathy and understanding. Quantitative research has shown that there is a statistically significant difference in transcultural features, rules and cultural patterns in the period of dying and death, depending on religion or personal belief, in which the degree of subjective identification with a religious belief or a personal belief plays an important role. According to the results of the investigation, it is obvious that de-taboooization of dying and death, as well as contact with the dying person and the deceased, greatly contributes to reconciliation with our own mortality. The results have also shown the presence of lower fears of death in women than men, which can also be seen from the results of the DAP-R questionnaire, which show that there is a statistically significant difference in the welcome acceptance of death by gender. The welcome acceptance to death is more expressed in believers than non-believers people. The focus group with nurses appears to have revealed the absence of cultural competencies that can greatly influence the quality of life of dying people and their families. The research study provides a comprehensive view of transcultural features, rules and cultural patterns in selected minorities in the Czech Republic. Based on data analysis, a nursing anamnesis, which can be an appropriate tool for the continual quality improvement of provided care, has been created. In view of the above mentioned, the research study is beneficial both for increasing theoretical knowledge of nursing and for clinical and community practice.

Em nossa vida diária, são utilizadas identidades digitais (IDDs) para acessar contas de e-mail, bancos e lojas virtuais, locais restritos, computadores compartilhados, e outros. Garantir que apenas usuários autorizados tenham o acesso permitido é um aspecto fundamental no desenvolvimento destas aplicações. Atualmente, os métodos de controle de acesso simples como senhas ou números de identificação pessoal não devem ser considerados suficientemente seguros, já que um impostor pode conseguir estas informações sem o conhecimento do usuário. Ainda, no caso de utilização de dispositivos físicos como cartões de identificação, estes podem ser roubados ou forjados. Para tornar estes sistemas mais confiáveis, técnicas de autenticação de identidades utilizando múltiplas verificações são propostas. A utilização de características biométricas surge como a alternativa mais confiável para tratar este problema, pois são, teoricamente, únicas para cada pessoa. Contudo, algumas características biométricas como a aparência facial podem variar com o tempo, implicando em um grande desafio para os sistemas de reconhecimento facial. Neste trabalho é combinado o acesso tradicional por senha com a análise da face para realizar a autenticação. Um método de aprendizagem supervisionada é apresentado e sua adaptação é baseada na melhora contínua dos modelos faciais, que são representados por misturas de gaussianas. Os resultados experimentais, obtidos sobre um conjunto de teste reduzido, são encorajadores, com 98% de identificação correta dos usuários e custo computacional relativamente baixo. Ainda, a comparação com um método apresentado na literatura indicou vantagens do método proposto quando usado como um pré-selecionador de faces. / In our daily life, we use digital identities (DIDs) to access e-mails, e-banks, e-shops, physical environments, shared computers, and so on. Guarantee that only authorized users are granted access is an important aspect in the development of such applications. Nowadays, the simple access control methods like passwords or personal identification numbers can not be considered secure enough, because an impostor can obtain and use these information without user knowledge. Also, physical devices like ID cards can be stolen. To make these systems more reliable, multimodal DID authentication techniques combining different verification steps are proposed. Biometric features appears as one of the most reliable alternatives to deal with this problem because, theoretically, they are unique for each person. Nevertheless, some biometric features like face appearances may change in time, posing a serious challenge for a face recognition system. In this thesis work, we use the traditional password access combined with human face analysis to perform the authentication task. An intuitive supervised appearance learning method is presented, and its adaptation is based on continuously improving face models represented using the Gaussian mixture modeling approach. The experimental results over a reduced test set show encouraging results, with 98% of the users correctly identified, with a relatively small computational effort. Still, the comparison with a method presented in the literature indicated advantages of the proposed method when used as a pre-selector of faces.

Neste trabalho, são apresentados a discussão de conceitos fundamentais, a metodologia e os resultados de simulações numéricas baseadas no Método de Volumes Finitos do escoamento de ar sobre algumas opções de configurações de turbinas eólicas do tipo Savonius, com e sem estatores, em operação e, também, em condições estáticas, como as encontradas nas partidas das mesmas. Comparam-se os resultados para diferentes domínios computacionais, bem como alternativas de discretização espacial e temporal, visando apresentar a influência desses sobre os valores obtidos e estabelecer os parâmetros computacionais adequados para a análise das turbinas em estudo. Nas simulações numéricas, desenvolvidas empregando o programa comercial Star-CCM+, a equação da continuidade e as equações de Navier-Stokes com médias de Reynolds são resolvidas, juntamente com as equações de um modelo de turbulência adequado, que é escolhido, para a obtenção dos campos de pressão e de velocidade do escoamento. Emprega-se um domínio contendo uma região com malha móvel, na qual o rotor é inserido. A cada simulação, a velocidade angular da região de malha móvel é especificada de maneira a variar a razão de velocidade de ponta do rotor. Através da integração das forças ocasionadas devido aos gradientes de pressão e das forças originadas pelo atrito viscoso sobre as pás do rotor eólico, obtém-se o coeficiente de torque em cada simulação. O torque e as forças atuantes no rotor também são obtidos de forma semelhante. Com esses dados, outros parâmetros como a potência e o coeficiente de potência são obtidos. Análises dos principais parâmetros de desempenho do rotor Savonius são realizadas e indicam uma boa concordância com resultados experimentais e de simulações numéricas realizadas por outros autores. Os resultados obtidos nas simulações apresentaram-se bastante representativos do fenômeno analisado. / This research work presents a discussion of basic concepts, the methodology and the results of numerical simulations based on Finite Volume Method for the air flow through some configuration options of the Savonius wind turbines, with and without stators, in operation, and also under static conditions, such as those found in the self starting. Results for different computational domains, as well as alternative spatial and temporal discretization are compared, in order to present the influence of these on the obtained values from the computational analysis of the turbines in study. In the numerical simulations, performed using the commercial software Star-CCM+, the equation of continuity and the Reynolds Averaged Navier-Stokes Equations were solved, together with the equations of a turbulence model appropriate, which is chosen, so that the fields of pressure and velocity could be found. It was used, in the calculations, a domain containing a region with a moving mesh, in which the rotor was inserted. In each simulation, the rotational rate of the moving mesh region was specified so as to vary the tip speed ratio of rotor. Through the integration of the forces arising due to the pressure gradients and the forces originated from the viscous friction on the wind rotor blades, the moment coefficient could be obtained in each simulation. The moment and forces acting on the rotor were also obtained similarly. With these data, other parameters such as the power and the power coefficient of the wind rotor could be obtained. Analysis of the principals performance parameters of the Savonius wind rotor were performed and indicated a good agreement with experimental results and numerical simulations performed by other authors. The simulations results are quite representative of the phenomenon analyzed.

abstract: Feature learning and the discovery of nonlinear variation patterns in high-dimensional data is an important task in many problem domains, such as imaging, streaming data from sensors, and manufacturing. This dissertation presents several methods for learning and visualizing nonlinear variation in high-dimensional data. First, an automated method for discovering nonlinear variation patterns using deep learning autoencoders is proposed. The approach provides a functional mapping from a low-dimensional representation to the original spatially-dense data that is both interpretable and efficient with respect to preserving information. Experimental results indicate that deep learning autoencoders outperform manifold learning and principal component analysis in reproducing the original data from the learned variation sources. A key issue in using autoencoders for nonlinear variation pattern discovery is to encourage the learning of solutions where each feature represents a unique variation source, which we define as distinct features. This problem of learning distinct features is also referred to as disentangling factors of variation in the representation learning literature. The remainder of this dissertation highlights and provides solutions for this important problem. An alternating autoencoder training method is presented and a new measure motivated by orthogonal loadings in linear models is proposed to quantify feature distinctness in the nonlinear models. Simulated point cloud data and handwritten digit images illustrate that standard training methods for autoencoders consistently mix the true variation sources in the learned low-dimensional representation, whereas the alternating method produces solutions with more distinct patterns. Finally, a new regularization method for learning distinct nonlinear features using autoencoders is proposed. Motivated in-part by the properties of linear solutions, a series of learning constraints are implemented via regularization penalties during stochastic gradient descent training. These include the orthogonality of tangent vectors to the manifold, the correlation between learned features, and the distributions of the learned features. This regularized learning approach yields low-dimensional representations which can be better interpreted and used to identify the true sources of variation impacting a high-dimensional feature space. Experimental results demonstrate the effectiveness of this method for nonlinear variation pattern discovery on both simulated and real data sets. / Dissertation/Thesis / Doctoral Dissertation Industrial Engineering 2016

Cette thèse porte sur la transformation de la voix d’un locuteur dans l’objectif d’indiquer la distance de celui-ci : une transformation en voix chuchotée pour indiquer une distance proche et une transformation en voix criée pour une distance plutôt éloignée. Nous effectuons dans un premier temps des analyses approfondies pour déterminer les paramètres les plus pertinentes dans une voix chuchotée et surtout dans une voix criée (beaucoup plus difficile). La contribution principale de cette partie est de montrer la pertinence des paramètres prosodiques dans la perception de l’effort vocal dans une voix criée. Nous proposons ensuite des descripteurs permettant de mieux caractériser les contours prosodiques. Pour la transformation proprement dite, nous proposons plusieurs nouvelles règles de transformation qui contrôlent de manière primordiale la qualité des voix transformées. Les résultats ont montré une très bonne qualité des voix chuchotées transformées ainsi que pour des voix criées pour des structures linguistiques relativement simples (CVC, CVCV, etc.). / This thesis focuses on speaker voice transformation in the aim to indicate the distance of it: a spokento-whispered voice transformation to indicate a close distance and a spoken-to-shouted voicetransformation for a rather far distance. We perform at first, in-depth analysis to determine mostrelevant features in whispered voices and especially in shouted voices (much harder). The maincontribution of this part is to show the relevance of prosodic parameters in the perception of vocaleffort in a shouted voice. Then, we propose some descriptors to better characterize the prosodiccontours. For the actual transformation, we propose several new transformation rules whichimportantly control the quality of transformed voice. The results showed a very good quality oftransformed whispered voices and transformed shouted voices for relatively simple linguisticstructures (CVC, CVCV, etc.).

Orientador: Ana Cláudia Ruggieri / Banca: Carlos Augusto de Miranda Gomide / Banca: Ricardo Andrade Reis / Resumo: O uso de pastagens é a alternativa mais econômica para a alimentação de ruminantes. É de suma importância o conhecimento das características morfogênicas e estruturais dos pastos, uma vez que as mesmas determinam o índice de área foliar do relvado, refletindo diretamente no pastejo pelos animais. O experimento foi conduzido no período de janeiro/2010 a junho/2010, com o objetivo de avaliar as características morfogênicas, estruturais e a dinâmica do perfilhamento do capim tanzânia em diferentes índices de área foliar residual (IAFr), sob pastejo rotacionado por cabras Anglonubiano. Os tratamentos consistiram de três IAF residual (0,8; 1,6 e 2,4). O delineamento experimental utilizado foi o de blocos completos casualizados, em arranjo de parcelas subdividas no tempo, com seis repetições e um total de 18 unidades experimentais. Para determinação do período de descanso foi adotado o critério de 95% de interceptação luminosa (condição de pré-pastejo). Foram avaliadas, no pré e pós pastejo, as alturas do dossel, IL, IAF, massa de forragem e as características morfogênicas e estruturais dos pastos. A altura média, IL e IAF dos dosséis no pré pastejo decresceram com a diminuição do IAFr. No pós pastejo, os IAFr 0,8 e 2,4 apresentaram maior e menor valor de IL, de 35,93 e 62,50%, respectivamente. No pré-pastejo os valores dos componentes de massa dos pastos mantidos sob IAFr 2,4 foram superiores (P<0,10), quando comparados aos pastos sob IAFr 0,8. Houve um aumento nos valores de massa verde seca (8696,85 kg/MS/ha) e massa seca total (12082,00 kg/MS/ha), conforme o ciclo de pastejo, devido principalmente à contribuição da massa seca de colmo (MSC). O período que compreendeu o último ciclo de pastejo teve a TApF 32% menor quando comparada aos primeiro e segundo ciclos de pastejo. O último ciclo de pastejo (Abr/Mai) apresentou TAlF inferior comparadas aos ... (Resumo completo, clicar acesso eletrônico abaixo) / Abstract: The use of pastures is the most economical way to feed ruminants. It is of paramount importance the knowledge of the morphogenetic and structural characteristics of the pasture, since they determine the leaf area index of the sward, is directly reflected by the grazing animals. The experiment was carried from janeiro2010 to June/2010, in order to evaluate the morphogenesis, structure and dynamics of tillering of Guinea grass in different residual LAI under rotational grazing by goats Anglo Nubian. The treatments consisted of three residual rLAI (0.8, 1.6 and 2.4). The experimental design was a complete randomized block, with a split plot arrangement, with six replication and a total of 18 plots. For the determination of rest period has been adopted the criterion of 95% light interception (under pre-grazing). There were evaluated before and after grazing, the sward heights, IL, LAI, herbage mass and the morphogenetic and structural characteristics of the pastures. The average height, canopy LAI and IL pre grazing decreased with decreasing rLAI. In the post grazing, the rLAI 0.8 and 2.4 showed higher and lower LI of 35.93 and 62.50% respectively. In the pre-grazing values of the components of the mass of swards under 2.4 rLAI were higher (P<0,10) when compared to pastures under rLAI 0.8. There was an increase in the values of green dry mass (8696,85 kg / DM / ha) and total dry mass (12082,00 kg / DM / ha) as the grazing cycle, mainly due to the contribution of the dry mass of stem (DMS). The period that included the last period of grazing had LAI 32% lower when compared to the first and second grazing cycles. The last grazing cycle (Apr / May) had markedly lower compared to previous cycles (Mar / Apr). The lifespan of leaves (VPD) was affected by treatments (rLAI) and the grazing season (P<0.10). The rLAI 2.4 resulted in longer life compared to rLAI 0.8 and 1.6. No significant rLAI ... (Complete abstract click electronic access below) / Mestre

Cette thèse modélise l’activité d’apprentissage dans les plateformes LMS (Learning Management System) en vue d’optimiser l’efficacité des utilisateurs. Ce projet est né d’une préoccupation pratique, à savoir comment faciliter l’utilisation des plateformes LMS chez les enseignants et apprenants des pays en voie de développement qui font face à la fracture numérique et éducative. Cette recherche aborde le problème de l’adaptabilité des LMS et suppose deux niveaux de modélisation : l’outil d’apprentissage et le contexte d’utilisation projeté. Pour traiter cette question d’adaptabilité, nous procédons par une double approche : l’analyse fonctionnelle des outils LMS et la réingénierie des interfaces utilisateurs. La première consiste à définir une approche d’analyse de l’activité d’enseignement-apprentissage dans les plateformes LMS. Ceci passe par une modélisation des situations courantes d’apprentissage et un croisement avec les fonctionnalités offertes dans les solutions LMS existantes. Ce travail préliminaire a permis de construire et proposer un modèle d’analyse utilisationnelle des plateformes que nous désignons méthode OCAPI fondé sur cinq catégories fonctionnelles : Organiser-Collaborer-Accompagner-Produire-Informer. La seconde approche s’inspire de la recherche fondamentale menée en amont et propose une réingénierie adaptative des LMS basée sur le contexte d’utilisation. Il s’agit d’un configurateur automatique embarqué qui adapte l’environnement de travail pour chaque usage et usager. Le prototype est articulé dans l’intention manifeste d’assurer une prise en main rapide des novices et se veut le moins contraignant possible du point de vue technologique. / The present research aims to model learning processes on Learning ManagementSystems (LMS) in a bid to maximize users’ efficiency. We came about this idea whilethinking over the possible ways of facilitating the use of LMS for teachers and learnersin countries affected by the digital divide. Drawing from that, the following question hasbeen stated: in a given learning context, how can we insert a Learning ManagementSystem that provides users with both easy handling and optimal using conditions? Thisissue raises the problem of LMS adaptability and suggests two levels of modeling: thelearning tool on one hand and the planned context of use on the other. To address thisissue of adaptability, we adopt a two-pronged approach including the functionalanalysis of LMS tools and the reengineering of user interfaces. The first step is todevelop an approach for the analysis of teaching and learning processes on LMS. Thisentails modeling common learning situations and cross-checking them with thefeatures available in LMS solutions. This preliminary work enabled to build a formalismfor LMS analysis which is referred to as the OCGPI approach (Organize-Collaborate-Guide-Produce-Inform). The second step proposes an adaptive reengineering of LMSbased on the context of use. This is namely an embedded configurator which adaptsthe working environment according to each use and each user. This tool aims at givingbeginners the possibility of acquainting themselves quickly with the virtual platform.