迴歸分析與類神經網路此兩種方法皆是預測領域上的主要工具。本論文嘗試在線性迴歸模式及非線性迴歸模式的條件下，隨機產生不同特性的資料以完整探討資料特性對迴歸分析與類神經網路之預測效果的影響。這些特性包括常態分配、偏態分配、不等變異、Michaelis-Menten關係模式及指數迴歸模式。 再者，我們使用區域搜尋法(local search methods)中的演化策略法(evolution strategies，ES)作為類神經網路的學習(learning)方法以提高其預測功能。我們稱這種類型的類神經網路為ESNN。 模擬結果顯示，ESNN確實可以取代常用來與迴歸分析做比較的倒傳遞類神經網路(back-propagation neural network，BPNN)，成為類神經網路的新選擇。針對不同特性的資料，我們建議：如果原始的資料適合以常態線性迴歸模式配適，則使用者可考慮使用迴歸方法做預測。如果原始的資料經由圖形分析或由檢定方法得知違反誤差項為均等變異之假設時，若能找到合適的權數，可使用加權最小平方法，但若權數難以決定時，則使用ESNN做預測。如果資料呈現韋伯偏態分佈時，可考慮使用ESNN或韋伯迴歸方法。資料適合以非線性迴歸模式做配適時，則選擇以ESNN做預測。 關鍵詞：迴歸分析，類神經網路，區域搜尋法，演化策略法類神經網路，倒傳遞類神經網路 / Both regression analysis and artificial neural networks are the main techniques for prediction. In this research, we tried to randomly generate different types of data, so as to completely explore the effect of data characteristics on the predictive performance of regression analysis and artificial neural networks. The data characteristics include normal distribution, skew distribution, unequal variances, Michaelis-Menten relationship model and exponential regression model. In addition, we used the evolution strategies, which is one of the local search methods for training artificial neural networks, to further improve its predictive performance. We name this type of artificial neural networks ESNN. Simulation studies indicate that ESNN could indeed replace BPNN to be the new choice of artificial neural networks. For different types of data, we commend that users can use regression analysis for their prediction if the original data is fit for linear regression model. When the residuals of the data are unequal variances, users can use weighted least squares if the optimal weights could be found. Otherwise, users can use ESNN. If the data is fit for weibull distribution, users can use ESNN or weibull regression. If the data is fit for nonlinear regression model, users can choose ESNN for the prediction. Keywords: Regression Analysis, Artificial Neural Networks, Local Search Methods, Evolution Strategies Neural Network (ESNN), Back-propagation Neural Network (BPNN)

L'hybridation des méthodes numériques est l'une des nombreuses pistes dans la recherche de la rapidité et de l'efficacité et de la précision d'une modélisation électromagnétique des structures complexes associant des parties de formes régulières de grandes dimensions électriques et des parties de formes complexes de dimensions plus modestes. Au lieu d'une seule formulation globale, on cherche à appliquer l'hybridation de plusieurs méthodes numériques notamment la méthode variationnelle multimodale (MVM), la méthode des éléments finis (FEM) et les réseaux de neurones artificiels. Un nouveau schéma hybride original qui combine la MVM et la FEM a été proposé pour caractériser une discontinuité complexe dans un guide d'onde rectangulaire. Les résultats obtenus tout en étant conformes aux résultats fournis par les simulateurs commerciaux et les résultats expérimentaux, apportent une amélioration sensible quant au temps de calcul. Le schéma hybride a été étendu pour la caractérisation des discontinuités complexes en cascade et appliqué à la conception de filtres micro-onde présentant des discontinuités complexes permettant ainsi un gain de temps très important. L'hybridation des réseaux de neurones artificiels et les méthodes modales a amélioré le temps de calcul pour l'analyse des discontinuités simples dans les guides d'onde rectangulaires ce qui a permis d'améliorer l'optimisation des filtres à guides d'ondes nervurés. / Hybridization of numerical methods is one inventive way in the research of the rapidity, the efficiency and the precision of the electromagnetic modeling of complex structures joining straight and large elements with complex and small ones. Instead of a global and unique formulation, we hybridize many numerical methods which are the modal methods, the finite element methods and the artificial neural networks. A novel computer- ided design (CAD) tool of complex passive microwave devices in rectangular waveguide technology is suggested. The multimodal variational method is applied to the full-wave description in the rectangular waveguides while the finite element analysis characterizes waves in the arbitrarily shaped discontinuities. The suggested hybrid approach is successfully applied to the full-wave analysis of complex discontinuities with great practical interest, thus improving CPU time and memory storage against several full-wave finite element method (FEM) based CAD tools. The proposed hybrid CAD tool is successfully extended to the design of filters with cascaded complex discontinuities. The hybridization of modal methods and the artificial neural networks improved the CPU time in the analysis of simple waveguide discontinuities which enhanced the optimization of rectangular ridged waveguide filters.

En apprentissage automatique, domaine qui consiste à utiliser des données pour apprendre une solution aux problèmes que nous voulons confier à la machine, le modèle des Réseaux de Neurones Artificiels (ANN) est un outil précieux. Il a été inventé voilà maintenant près de soixante ans, et pourtant, il est encore de nos jours le sujet d'une recherche active. Récemment, avec l'apprentissage profond, il a en effet permis d'améliorer l'état de l'art dans de nombreux champs d'applications comme la vision par ordinateur, le traitement de la parole et le traitement des langues naturelles. La quantité toujours grandissante de données disponibles et les améliorations du matériel informatique ont permis de faciliter l'apprentissage de modèles à haute capacité comme les ANNs profonds. Cependant, des difficultés inhérentes à l'entraînement de tels modèles, comme les minima locaux, ont encore un impact important. L'apprentissage profond vise donc à trouver des solutions, en régularisant ou en facilitant l'optimisation. Le pré-entraînnement non-supervisé, ou la technique du ``Dropout'', en sont des exemples. Les deux premiers travaux présentés dans cette thèse suivent cette ligne de recherche. Le premier étudie les problèmes de gradients diminuants/explosants dans les architectures profondes. Il montre que des choix simples, comme la fonction d'activation ou l'initialisation des poids du réseaux, ont une grande influence. Nous proposons l'initialisation normalisée pour faciliter l'apprentissage. Le second se focalise sur le choix de la fonction d'activation et présente le rectifieur, ou unité rectificatrice linéaire. Cette étude a été la première à mettre l'accent sur les fonctions d'activations linéaires par morceaux pour les réseaux de neurones profonds en apprentissage supervisé. Aujourd'hui, ce type de fonction d'activation est une composante essentielle des réseaux de neurones profonds. Les deux derniers travaux présentés se concentrent sur les applications des ANNs en traitement des langues naturelles. Le premier aborde le sujet de l'adaptation de domaine pour l'analyse de sentiment, en utilisant des Auto-Encodeurs Débruitants. Celui-ci est encore l'état de l'art de nos jours. Le second traite de l'apprentissage de données multi-relationnelles avec un modèle à base d'énergie, pouvant être utilisé pour la tâche de désambiguation de sens. / Machine learning aims to leverage data in order for computers to solve problems of interest. Despite being invented close to sixty years ago, Artificial Neural Networks (ANN) remain an area of active research and a powerful tool. Their resurgence in the context of deep learning has led to dramatic improvements in various domains from computer vision and speech processing to natural language processing. The quantity of available data and the computing power are always increasing, which is desirable to train high capacity models such as deep ANNs. However, some intrinsic learning difficulties, such as local minima, remain problematic. Deep learning aims to find solutions to these problems, either by adding some regularisation or improving optimisation. Unsupervised pre-training or Dropout are examples of such solutions. The two first articles presented in this thesis follow this line of research. The first analyzes the problem of vanishing/exploding gradients in deep architectures. It shows that simple choices, like the activation function or the weights initialization, can have an important impact. We propose the normalized initialization scheme to improve learning. The second focuses on the activation function, where we propose the rectified linear unit. This work was the first to emphasise the use of linear by parts activation functions for deep supervised neural networks, which is now an essential component of such models. The last two papers show some applications of ANNs to Natural Language Processing. The first focuses on the specific subject of domain adaptation in the context of sentiment analysis, using Stacked Denoising Auto-encoders. It remains state of the art to this day. The second tackles learning with multi-relational data using an energy based model which can also be applied to the task of word-sense disambiguation.

Esta pesquisa visa a análise da contribuição de cinco variáveis de entrada e a otimização do desempenho termo-hidráulico de trocadores de calor com venezianas combinados com geradores de vórtices delta-winglets. O desempenho termohidráulico de duas geometrias distintas, aqui nomeadas por GEO1 e GEO2, foram avaliadas. Smoothing Spline ANOVA foi usado para avaliar a contribuição dos parâmetros de entrada na transferência de calor e perda de carga. Considerando aplicação automotiva, foram investigados números de Reynolds iguais a 120 e 240, baseados no diâmetro hidráulico. Os resultados indicaram que o ângulo de venezianas é o maior contribuidor para o aumento do fator de atrito para GEO1 e GEO2, para ambos os números de Reynolds. Para o número de Reynolds menor, o parâmetro mais importante em termos de transferência de calor foi o ângulo das venezianas para ambas as geometrias. Para o número de Reynolds maior, o ângulo de ataque dos geradores de vórtices posicionados na primeira fileira é o maior contribuidor para a tranfesferência de calor, no caso da geometria GEO1, enquanto que o ângulo de ataque dos geradores de vórtices na primeira fileira foi tão importante quanto os ângulos das venezianas para a geometria GEO2. Embora as geometrias analisadas possam ser consideradas como técnicas compostas de intensificação da transferência de calor, não foram observadas interações relevantes entre ângulo de venezianas e parâmetros dos geradores de vórtices. O processo de otimização usa NSGA-II (Non-Dominated Sorting Genetic Algorithm) combinado com redes neurais artificiais. Os resultados mostraram que a adição dos geradores de vórtices em GEO1 aumentaram a transferência de calor em 21% e 23% com aumentos na perda de carga iguais a 24,66% e 36,67% para o menor e maior números de Reynolds, respectivamente. Para GEO2, a transferência de calor aumentou 13% e 15% com aumento na perda de carga de 20,33% e 23,70%, para o menor e maior número de Reynolds, respectivamente. As soluções otimizadas para o fator de Colburn mostraram que a transferência de calor atrás da primeira e da segunda fileiras de geradores de vórtices tem a mesma ordem de magnitude para ambos os números de Reynolds. Os padrões de escoamento e as características de transferência de calor das soluções otimizadas apresentaram comportamentos vi particulares, diferentemente daqueles encontrados quando as duas técnicas de intensificação de transferência de calor são aplicadas separadamente. / This doctoral thesis focuses on screening analysis of five input parameters and heat transfer and pressure drop optimization of flat-tube multi-louvered fin heat exchangers combined with delta-winglet vortex generators. The thermal-hydraulic performance of two distinct geometries, GEO1 and GEO2, were evaluated. Smoothing Spline ANOVA was used to evaluate the contribution of the input parameters such as louver angle, angle of attack of the delta-winglet and streamwise position of the delta-winglet on heat transfer and pressure drop. Taking the automotive application into account, Reynolds numbers of 120 and 240, based on hydraulic diameter, were investigated. The results indicated that the louver angle is the main contributor to increase the Friction factor for GEO1 and GEO2 for both Reynolds numbers. For the lower Reynolds number, the most important heat transfer parameter was the louver angle for both geometries, while at the higher Reynolds number, the angles of attack of the first row of delta-winglets mostly contributed to GEO1, and the angle of attack of the first row of delta-winglets was as important as the louver angle for GEO2. Although those specific geometries can be considered a kind of compound enhancement technique, relevant interactions were not verified between louvers and delta-winglet vortex generators parameters. The surrogatebased optimization procedure uses NSGA-II method (Non-Dominated Sorting Genetic Algorithm) combined with artificial neural networks. The results showed that the addition of DWLs on GEO1 increased the heat transfer of 21.27% and 23.52% with associated pressure loss increasing of 24.66% and 36.67% for the lower and the higher Reynolds numbers, respectively. For GEO2, the heat transfer was increased 13.48% and 15.67% with an increase of the pressure drop of 20.33% and 23.70%, for the lower and the higher Reynolds numbers, respectively. The optimized solutions for the Colburn factor showed that heat transfer behind the second row of deltawinglets has the same order of magnitude of that behind the first row, for both Reynolds numbers. The flow patterns and heat transfer characteristics from optimized solutions presented some particular behavior, differently from the findings when those two heat transfer enhancement techniques are applied separately.

O objetivo deste trabalho foi desenvolver um modelo para reconhecer e reproduzir padrões de viagens encadeadas a pé. O processo de modelagem foi conduzido através da aplicação das técnicas das Redes Neurais Artificiais (RNAs), utilizando-se de uma rede estática MLP e de rede dinâmica Elman. A análise do desempenho do modelo foi baseada nos dados de uma pesquisa de Origem-Destino realizada, em 1987, pelo METRÔ-SP na Região Metropolitana de São Paulo. Na modelagem foi fixado o modo de viagem a pé, e, na abordagem seqüencial, padrões de viagens individuais foram representados em termos de dois componentes: duração da viagem e tipo de atividades. A análise foi realizada partindo da classificação geral e específica para cada segmento do encadeamento de viagens, o que permitiu a comparação dos resultados entre padrões de viagens observados e os reproduzidos pelas redes. Na classificação geral, cinco dos padrões previstos com maior freqüência pelas RNAs representaram em média 58,9% dos indivíduos no conjunto de dados usado para testar o desempenho do modelo. Para o vetor de duas e quatro viagens, as redes neurais reproduziram 50% das durações de viagem e 90% das atividades, tais como Trabalho e Escola. Embora esses resultados não pareçam muito robustos, não significa que eles estejam errados. As porcentagens acima representam a probabilidade de uma pessoa realizar viagens com aquelas durações ou tipo de atividades. / The main objective of this work was to develop a model for recognizing and reproduzing trip-chaining patterns by walk. The process of modeling was conducted applying the techniques of Artificial Neural Networks (ANNs), by using one of the static networks MLP and the Elman dynamic network. The analysis of the performance of the model was based on the origin-destination home-interview survey carried out by METRÔ-SP in São Paulo Metropolitan Area in 1987. The mode of trip by walk was fixed in the model, and, in the sequential approach, individual travel patterns were represented in terms of two components: trip duration and activity type. The analysis was accomplished starting from the general and specific classifications for each segment of the chained trips, which allowed the comparison of the results between the observed travel patterns and reproduced ones through ANNs. In general classification, 5 of the patterns most frequently predicted by the ANNs represented 58.9% of the individuals in the dataset used for testing the model performance. For the vectors of two and four trips, the neural networks reproduced 50% of trip durations and 90% of the activities, such as work and school. Although those results seem not so robust, it does not mean that they are wrong. The percentages above represent the probability of a person making trips with those durations or type of activities.

Moderne Präzisionsgartenbaulicheproduktion schließt hoch technifizierte Gewächshäuser, deren Einsatz in großem Maße von der Qualität der Sensorik- und Regelungstechnik abhängt, mit ein. Zu den Regelungsstrategien gehören unter anderem Methoden der Künstlichen Intelligenz, wie z.B. Künstliche Neuronale Netze (KNN, aus dem Englischen). Die vorliegende Arbeit befasst sich mit der Eignung KNN-basierter Modelle als Bauelemente von Klimaregelungstrategien in Gewächshäusern. Es werden zwei Modelle vorgestellt: Ein Modell zur kurzzeitigen Voraussage des Gewächshausklimas (Lufttemperatur und relative Feuchtigkeit, in Minuten-Zeiträumen), und Modell zur Einschätzung von phytometrischen Signalen (Blatttemperatur, Transpirationsrate und Photosyntheserate). Eine Datenbank, die drei Kulturjahre umfasste (Kultur: Tomato), wurde zur Modellbildung bzw. -test benutzt. Es wurde festgestellt, dass die ANN-basierte Modelle sehr stark auf die Auswahl der Metaparameter und Netzarchitektur reagieren, und dass sie auch mit derselben Architektur verschiedene Kalkulationsergebnisse liefern können. Nichtsdestotrotz, hat sich diese Art von Modellen als geeignet zur Einschätzung komplexer Pflanzensignalen sowie zur Mikroklimavoraussage erwiesen. Zwei zusätzliche Möglichkeiten zur Erstellung von komplexen Simulationen sind in der Arbeit enthalten, und zwar zur Klimavoraussage in längerer Perioden und zur Voraussage der Photosyntheserate. Die Arbeit kommt zum Ergebnis, dass die Verwendung von KNN-Modellen für neue Gewächshaussteuerungstrategien geeignet ist, da sie robust sind und mit der Systemskomplexität gut zurechtkommen. Allerdings muss beachtet werden, dass Probleme und Schwierigkeiten auftreten können. Diese Arbeit weist auf die Relevanz der Netzarchitektur, die erforderlichen großen Datenmengen zur Modellbildung und Probleme mit verschiedenen Zeitkonstanten im Gewächshaus hin. / One facet of the current developments in precision horticulture is the highly technified production under cover. The intensive production in modern greenhouses heavily relies on instrumentation and control techniques to automate many tasks. Among these techniques are control strategies, which can also include some methods developed within the field of Artificial Intelligence. This document presents research on Artificial Neural Networks (ANN), a technique derived from Artificial Intelligence, and aims to shed light on their applicability in greenhouse vegetable production. In particular, this work focuses on the suitability of ANN-based models for greenhouse environmental control. To this end, two models were built: A short-term climate prediction model (air temperature and relative humidity in time scale of minutes), and a model of the plant response to the climate, the latter regarding phytometric measurements of leaf temperature, transpiration rate and photosynthesis rate. A dataset comprising three years of tomato cultivation was used to build and test the models. It was found that this kind of models is very sensitive to the fine-tuning of the metaparameters and that they can produce different results even with the same architecture. Nevertheless, it was shown that ANN are useful to simulate complex biological signals and to estimate future microclimate trends. Furthermore, two connection schemes are proposed to assemble several models in order to generate more complex simulations, like long-term prediction chains and photosynthesis forecasts. It was concluded that ANN could be used in greenhouse automation systems as part of the control strategy, as they are robust and can cope with the complexity of the system. However, a number of problems and difficulties are pointed out, including the importance of the architecture, the need for large datasets to build the models and problems arising from different time constants in the whole greenhouse system.

Depuis son introduction en 1982, la carte auto-organisatrice de Kohonen (Self-Organizing Map : SOM) a prouvé ses capacités de classification et visualisation des données multidimensionnelles dans différents domaines d’application. Les implémentations matérielles de la carte SOM, en exploitant le taux de parallélisme élevé de l’algorithme de Kohonen, permettent d’augmenter les performances de ce modèle neuronal souvent au détriment de la flexibilité. D’autre part, la flexibilité est offerte par les implémentations logicielles qui quant à elles ne sont pas adaptées pour les applications temps réel à cause de leurs performances temporelles limitées. Dans cette thèse nous avons proposé une architecture matérielle distribuée, adaptable, flexible et extensible de la carte SOM à base de NoC dédiée pour une implantation matérielle sur FPGA. A base de cette approche, nous avons également proposé une architecture matérielle innovante d’une carte SOM à structure croissante au cours de la phase d’apprentissage / Since its introduction in 1982, Kohonen’s Self-Organizing Map (SOM) showed its ability to classify and visualize multidimensional data in various application fields. Hardware implementations of SOM, by exploiting the inherent parallelism of the Kohonen algorithm, allow to increase the overall performances of this neuronal network, often at the expense of the flexibility. On the other hand, the flexibility is offered by software implementations which on their side are not suited for real-time applications due to the limited time performances. In this thesis we proposed a distributed, adaptable, flexible and scalable hardware architecture of SOM based on Network-on-Chip (NoC) designed for FPGA implementation. Moreover, based on this approach we also proposed a novel hardware architecture of a growing SOM able to evolve its own structure during the learning phase

Cherenkov-Teleskope sind in der Lage, das schwache Cherenkovlicht aus Teilchenschauern zu detektieren, die von kosmischen Teilchen mit Energien von ca. 100 GeV bis 100 TeV in der Erdatmosphäre initiiert werden. Das Ziel ist die Detektion von Cherenkovlicht aus Schauern, die von Gammastrahlen erzeugt wurden, der größte Teil der Schauer stammt jedoch von geladenen Teilchen. Im Jahr 2012 wurde das H.E.S.S.-Observatorium in Namibia, bis dahin bestehend aus vier Teleskopen mit 100 m²-Spiegeln, um ein fünftes Teleskop mit einer Spiegelfläche von ca. 600 m² ergänzt. Aufgrund der großen Spiegelfläche besitzt dieses Teleskop die niedrigste Energieschwelle aller Teleskope dieser Art. In dieser Dissertation wird ein schneller Algorithmus namens MonoReco präsentiert, der grundlegende Eigenschaften der Gammastrahlen wie ihre Energien und Richtungen rekonstruieren kann. Dieser Algorithmus kann weiterhin unterscheiden, ob Schauer von Gammastrahlen oder von geladenen Teilchen der kosmischen Strahlung initiiert wurden. Diese Aufgaben werden mit mithilfe von künstlichen neuronalen Netzwerken erfüllt, welche ausschließlich die Momente der Intensitätsverteilungen in der Kamera des neuen Teleskops analysieren. Eine Energieschwelle von 59 GeV und Richtungsauflösungen von 0.1°-0.3° werden erreicht. Das Energiebias liegt bei wenigen Prozent, die Energieauflösung bei 20-30%. Unter anderem mit dem MonoReco-Algorithmus wurden Daten, die in der Zeit um das Periastron des Binärsystems PSR B1259-63/LS 2883 im Jahre 2014 genommen wurden, analysiert. Es handelt sich hierbei um einen Neutronenstern, der sich in einem 3,4-Jahres-Orbit um einen massereichen Stern mit einer den Stern umgebenden Scheibe aus Gas und Plasmen befindet. Zum ersten Mal konnte H.E.S.S. das Gammastrahlenspektrum dieses Systems bei Energien unterhalb von 200 GeV messen. Weiterhin wurde bei erstmaligen Beobachtungen zur Zeit des Periastrons ein lokales Flussminimum gemessen. Sowohl vor dem ersten als auch nach dem zweiten Transit des Neutronensterns durch die Scheibe wurden hohe Flüsse gemessen. Im zweiten Fall wurden Beobachtungen erstmals zeitgleich mit dem Fermi-LAT-Experiment durchgeführt, das wiederholt sehr hohe Flüsse in diesem Teil des Orbits messen konnte. Ein Vergleich der gemessenen Flüsse mit Vorhersagen eines leptonischen Modells zeigt gute Übereinstimmungen. / Cherenkov telescopes can detect the faint Cherenkov light emitted by air showers that were initiated by cosmic particles with energies between approximately 100 GeV and 100 TeV in the Earth's atmosphere. Aiming for the detection of Cherenkov light emitted by gamma ray-initiated air showers, the vast majority of all detected showers are initiated by charged cosmic rays. In 2012 the H.E.S.S. observatory, until then comprising four telescopes with 100 m² mirrors each, was extended by adding a much larger fifth telescope with a very large mirror area of 600 m². Due to the large mirror area, this telescope has the lowest energy threshold of all telescopes of this kind. In this dissertation, a fast algorithm called MonoReco is presented that can reconstruct fundamental properties of the primary gamma rays like their direction or their energy. Furthermore, this algorithm can distinguish between air showers initiated either by gamma rays or by charged cosmic rays. Those tasks are accomplished with the help of artificial neural networks, which analyse moments of the intensity distributions in the camera of the new telescope exclusively. The energy threshold is 59 GeV and angular resolutions of 0.1°-0.3° are achieved. The energy reconstruction bias is at the level of a few percent, the energy resolution is at the level of 20-30%. Data taken around the 2014 periastron passage of the gamma-ray binary PSR B1259-63/LS 2883 were analysed with, among others, the MonoReco algorithm. This binary system comprises a neutron star in a 3.4 year orbit around a massive star with a circumstellar disk consisting of gas and plasma. For the first time the gamma-ray spectrum of this system could be measured by H.E.S.S. down to below 200 GeV. Furthermore, a local flux minimum could be measured during unprecedented measurements at the time of periastron. High fluxes were measured both before the first and after the second transit of the neutron star through the disk. In the second case measurements could be performed for the first time contemporaneously with the Fermi-LAT experiment, which has repeatedly detected very high fluxes at this part of the orbit. A good agreement between measured fluxes and predictions of a leptonic model is found.

Este trabalho apresenta um sistema computacional que utiliza técnicas de Aprendizado de Máquina (AM) para auxiliar o diagnóstico oftalmológico. Trata-se de um sistema de medidas objetivas e automáticas dos principais vícios de refração ocular, astigmatismo, hipermetropia e miopia. O sistema funcional desenvolvido aplica técnicas convencionais de processamento a imagens do olho humano fornecidas por uma técnica de aquisição chamada Hartmann-Shack (HS), ou Shack-Hartmann (SH), com o objetivo de extrair e enquadrar a região de interesse e remover ruídos. Em seguida, vetores de características são extraídos dessas imagens pela técnica de transformada wavelet de Gabor e, posteriormente, analisados por técnicas de AM para diagnosticar os possíveis vícios refrativos presentes no globo ocular representado. Os resultados obtidos indicam a potencialidade dessa abordagem para a interpretação de imagens de HS de forma que, futuramente, outros problemas oculares possam ser detectados e medidos a partir dessas imagens. Além da implementação de uma nova abordagem para a medição dos vícios refrativos e da introdução de técnicas de AM na análise de imagens oftalmológicas, o trabalho contribui para a investigação da utilização de Máquinas de Vetores Suporte e Redes Neurais Artificiais em sistemas de Entendimento/Interpretação de Imagens (Image Understanding). O desenvolvimento deste sistema permite verificar criticamente a adequação e limitações dessas técnicas para a execução de tarefas no campo do Entendimento/Interpretação de Imagens em problemas reais. / This work presents a computational system that uses Machine Learning (ML) techniques to assist in ophthalmological diagnosis. The system developed produces objective and automatic measures of ocular refraction errors, namely astigmatism, hypermetropia and myopia from functional images of the human eye acquired with a technique known as Hartmann-Shack (HS), or Shack-Hartmann (SH). Image processing techniques are applied to these images in order to remove noise and extract the regions of interest. The Gabor wavelet transform technique is applied to extract feature vectors from the images, which are then input to ML techniques that output a diagnosis of the refractive errors in the imaged eye globe. Results indicate that the proposed approach creates interesting possibilities for the interpretation of HS images, so that in the future other types of ocular diseases may be detected and measured from the same images. In addition to implementing a novel approach for measuring ocular refraction errors and introducing ML techniques for analyzing ophthalmological images, this work investigates the use of Artificial Neural Networks and Support Vector Machines (SVMs) for tasks in Image Understanding. The description of the process adopted for developing this system can help in critically verifying the suitability and limitations of such techniques for solving Image Understanding tasks in \"real world\" problems.

Informações sobre as características funcionais e estruturais da unidade motora (UM) são altamente relevantes em investigações fisiológicas e nos estudos clínicos das disfunções neuromusculares. A eletromiografia (EMG) é um método adequado para obtenção dessas informações. Entretanto, devido à dificuldade na separação da atividade individual de uma unidade motora das outras que estão simultaneamente ativas, seu uso em clínica prática se dá comumente através de métodos invasivos, empregando eletrodos de agulha ou fios implantados. Apesar da EMG de superfície ser não-invasiva e, portanto mais apropriada para aplicações clínicas, não é usada em clínica porque não há até o presente um método satisfatório para decomposição do sinal EMG de superfície. Um EMG de superfície é muito mais difícil de decompor devido a significante superposição dos Potenciais de Ação das UMs (MUAPs) e a relação sinal-ruído relativamente baixa, se comparada aos métodos invasivos. Defendemos que a separação da atividade individual das UMs pode ser feita de modo não-invasivo aliando-se técnicas de aquisição altamente especializadas com técnicas usadas em reconhecimento de padrões. Desenvolvemos um método para decomposição de EMGs de superfície, a partir do qual foi possível extrair características relevantes das UMs, que permitem seu uso em avaliação e diagnóstico de desordens neuromusculares. Em nossa abordagem, o sinal EMG é inicialmente captado sob contração isométrica fraca usando eletrodos desuperfície. O sinal EMG bruto passa em seguida por um filtro Diferencial Passa-Baixas Ponderado (DPBP) em série com um detector de picos, que detecta os picos de MUAPs e extrai suas formas de onda. Na sequência, o conjunto de MUAPs extraído é classificado por uma rede neural SOM, e os MUAPs agrupados pela similaridade de suas formas de onda. No próximo passo a informação temporal dos disparos é checada, eliminando possíveis erros de classificação, e finalmente os Trens de MUAPs (MUAPTs) das UMs individuais são reconstituídos do EMG original. As estatísticas de disparos (IPI) bem como as formas de ondas dos MUAPs das respectivas UMs são então extraídas e armazenadas para estudos posteriores. Resultados preliminares obtidos com EMGs normais e patológicos, extraídos de membros superiores sob contração fraca, indicam que, o método mostrou-se apto a decompor EMGs de superfícies, além de potencial para aplicações em estudos clínicos não-invasivos de disfunções neuromusculares.Informações sobre as características funcionais e estruturais da unidade motora (UM) são altamente relevantes em investigações fisiológicas e nos estudos clínicos das disfunções neuromusculares. A eletromiografia (EMG) é um método adequado para obtenção dessas informações. Entretanto, devido à dificuldade na separação da atividade individual de uma unidade motora das outras que estão simultaneamente ativas, seu uso em clínica prática se dá comumente através de métodos invasivos, empregando eletrodos de agulha ou fios implantados. Apesar da EMG de superfície ser não-invasiva e, portanto mais apropriada para aplicações clínicas, não é usada em clínica porque não há até o presente um método satisfatório para decomposição do sinal EMG de superfície. Um EMG de superfície é muito mais difícil de decompor devido a significante superposição dos Potenciais de Ação das UMs (MUAPs) e a relação sinal-ruído relativamente baixa, se comparada aos métodos invasivos. Defendemos que a separação da atividade individual das UMs pode ser feita de modo não-invasivo aliando-se técnicas de aquisição altamente especializadas com técnicas usadas em reconhecimento de padrões. Desenvolvemos um método para decomposição de EMGs de superfície, a partir do qual foi possível extrair características relevantes das UMs, que permitem seu uso em avaliação e diagnóstico de desordens neuromusculares. Em nossa abordagem, o sinal EMG é inicialmente captado sob contração isométrica fraca usando eletrodos desuperfície. O sinal EMG bruto passa em seguida por um filtro Diferencial Passa-Baixas Ponderado (DPBP) em série com um detector de picos, que detecta os picos de MUAPs e extrai suas formas de onda. Na sequência, o conjunto de MUAPs extraído é classificado por uma rede neural SOM, e os MUAPs agrupados pela similaridade de suas formas de onda. No próximo passo a informação temporal dos disparos é checada, eliminando possíveis erros de classificação, e finalmente os Trens de MUAPs (MUAPTs) das UMs individuais são reconstituídos do EMG original. As estatísticas de disparos (IPI) bem como as formas de ondas dos MUAPs das respectivas UMs são então extraídas e armazenadas para estudos posteriores. Resultados preliminares obtidos com EMGs normais e patológicos, extraídos de membros superiores sob contração fraca, indicam que, o método mostrou-se apto a decompor EMGs de superfícies, além de potencial para aplicações em estudos clínicos não-invasivos de disfunções neuromusculares. / Information on the functional and structural characteristics of the motor unit (MU) they are highly important in physiologic investigations and in the clinical studies of the neuromuscular dysfunctions. The electromyography (EMG) it is an appropriate method for obtaining of that information. However, due to the difficulty in the separation of the individual activity of a motor unit of the another that are simultaneously active, your use in practical clinic happen commonly through methods invasive, employing needle electrodes or implanted threads. In spite of surface EMG to be non-invasive and, therefore more appropriate for clinical applications, it is not used at clinic because there is not until the present a satisfactory method for decomposition of the surface EMG sign. A surface EMG is much more difficult of decomposing due to significant overlap of the Motor Unit Action Potentials (MUAPs) and the relationship sign-noise relatively low, if compared to the invasive methods. We defended that the separation of the individual activity of MUs can be made in way non-invasive allying highly specialized acquisition techniques with techniques used in recognition of patterns. We developed a method for decomposition of surface EMGs, starting from which was possible to extract important characteristics of MUs, which allow your use in evaluation and diagnosis of neuromuscular disorders. In our approach, the sign EMG is captured initially under weak isometriccontraction using surface electrodes. The sign EMG raw raisin soon after for a Biased Low-Pass Differential filter (BLPD) in series with a detector of peaks, that detects the peaks of MUAPs and it extracts your wave forms. In the sequence, a SOM neural network classifies the set of extracted MUAPs, and MUAPs are clustered by the similarity in your wave shape. In the next step the temporal information of the discharges is checked, eliminating possible classification mistakes, and finally the MUAPs Trains (MUAPTs) of individual MUs they are reconstituted of original EMG. The statistics of discharges (IPI) as well as the forms of waves of MUAPs of respective MUs are then extracted and stored for subsequent studies. Results preliminaries obtained with normal and pathological EMGs, extracted of superior members under weak contraction, they indicate that, the method was shown capable to decompose surfaces EMGs, besides potential for applications in clinical studies non-invasive of neuromuscular dysfunctions.