Garantir a qualidade da carne bovina produzida no Brasil tem sido uma preocupação dos produtores, pois contribui para aumentar a exportação e o consumo interno do produto. Por isso, tem-se pesquisado novos métodos que analisam e garantam a qualidade da carne, de forma rápida, eficiente e não destrutiva. A ressonância magnética nuclear (RMN) tem se destacado como uma das técnicas de controle de qualidade de carne. Neste trabalho as redes neurais artificiais estão sendo utilizadas para o reconhecimento de padrões dos dados de ressonância magnética nuclear oriundos de carne bovina. Mais especificamente, os respectivos dados têm sido utilizados por uma rede perceptron multicamadas para a extração de características da carne bovina, possibilitando a classificação do grupo genético e do sexo dos animais a partir de uma amostra da referida carne. Os resultados dos experimentos são também apresentados para ilustrar o desempenho da abordagem proposta. / Guaranteeing the quality of the bovine meat produced in Brazil has been a concern of the producers because it contributes to increase the export and the domestic consumption of the product. Therefore, new methods have been researched that analyze and guarantee the quality of the meat in a fast, efficient and non destructive way. Nuclear magnetic resonance (NMR) has been highlighted as one of the techniques of meat quality control. In this work study artificial neural networks are being used for pattern recognition from data obtained by the resonance equipment, originating from bovine meat. More specifically, the respective data have been used by a multilayer perceptron network for extraction of bovine meat characteristics, making possible the classification of both genetic group and animal sex starting from a single meat sample. Several results of experimental tests are also presented to illustrate the performance of the proposed approach.

We present an example-based learning approach for locating vertical frontal views of human faces in complex scenes. The technique models the distribution of human face patterns by means of a few view-based "face'' and "non-face'' prototype clusters. At each image location, the local pattern is matched against the distribution-based model, and a trained classifier determines, based on the local difference measurements, whether or not a human face exists at the current image location. We provide an analysis that helps identify the critical components of our system.

Prediction of protein secondary structure from primary sequence of amino acids is a very challenging task, and the problem has been approached from several angles. Proteins have many different biological functions; they may act as enzymes or as building blocks (muscle fibers) or may have transport function (e.g., transport of oxygen). The three-dimensional protein structure determines the functional properties of the protein. A lot of interesting work has been done on this problem, and over the last 10 to 20 years the methods have gradually improved in accuracy. In this dissertation we investigate several techniques for predicting the protein secondary structure. The prediction is carried out mainly using pattern classification techniques such as neural networks, genetic algorithms, simulated annealing. Each individual algorithm may work well in certain situations but fails in others. Capitalizing on the positive decisions can be achieved by forcing the various methods to collaborate to reach a unified consensus based on their previous performances. The process of combining classifiers is called decision fusion. The various decision fusion techniques such as the committee method, correlation method and the Bayesian inference methods to fuse the solutions from various approaches and to get better prediction accuracy are thoroughly explored in this dissertation. The RS126 data set was used for training and testing purposes. The results of applying pattern classification algorithms along with decision fusion techniques showed improvement in the prediction accuracy compared to that of prediction by neural networks or pattern classification algorithms individually or combined with neural networks. This research has shown that decision fusion techniques can be used to obtain better protein secondary structure prediction accuracy.

Classification of migraineur and healthy subjects using statistical pattern classifiers on functional Near Infrared Spectroscopy (NIRS) data is the main purpose of this study. Also a statistical comparison between trials that have different type of classifiers, classifier settings and feature sets is done. Features are extracted from raw light measurement data acquired with NIRS device, namely Niroxcope, during two separate previous studies, using Modified Beer-Lambert Law. After feature extraction, Na&iuml / ve Bayes classifier and k Nearest Neighbor classifier are utilized with and with-out Principal Component Analysis in separate trials. Results obtained are compared within each other using statistical hypothesis tests namely Mc Nemar and Cochran Q.

Developement of a procedure to calculate intracranial pressure by means of arterial blood pressure and blood flow velocity in a big cerebral artery. Methods of systems theory are used. / Entwicklung eines Verfahrens zur Berechnung des Hirndrucks aus dem Bludrucksignal und der Blutströmungsgeschwindigkeit in einer großen Hirnarterie. Es werden Methoden der Systemtheorie verwendet.

Die Berechnung der Netzverluste in Niederspannungsnetzen kann nur über Umwege erfolgen, da viele Netzbetreiber keine digitalisierten Daten ihrer Netze besitzen. Hier wird ein Ansatz zur Ermittlung der technischen Verluste anhand vorhandener Netzdaten beschrieben. / The major part of network operator of low voltage nets do not have digitised data of their nets. That’s why net losses must be calculated indirectly. This article describes an approach for getting results in a direct way.

This thesis examines the application of a Support Vector Machine (SVM) classifier to automatically detect sleep and quiet wake (rest) behavior in mice from pressure signals on their cage floor. Previous work employed Neural Networks (NN) and Linear Discriminant Analysis (LDA) to successfully detect sleep and wake behaviors in mice. Although the LDA was successful in distinguishing between the sleep and wake behaviors, it has several limitations, which include the need to select a threshold and difficulty separating additional behaviors with subtle differences, such as sleep and rest. The SVM has advantages in that it offers greater degrees of freedom than the LDA for working with complex data sets. In addition, the SVM has direct methods to limit overfitting for the training sets (unlike the NN method). This thesis develops an SVM classifier to characterize the linearly non separable sleep and rest behaviors using a variety of features extracted from the power spectrum, autocorrelation function, and generalized spectrum (autocorrelation of complex spectrum). A genetic algorithm (GA) optimizes the SVM parameters and determines a combination of 5 best features. Experimental results from over 9 hours of data scored by human observation indicate 75% classification accuracy for SVM compared to 68% accuracy for LDA.

Pattern recognition in urban areas is one of the most challenging issues in classifying satellite remote sensing data. Parametric pixel-by-pixel classification algorithms tend to perform poorly in this context. This is because urban areas comprise a complex spatial assemblage of disparate land cover types - including built structures, numerous vegetation types, bare soil and water bodies. Thus, there is a need for more powerful spectral pattern recognition techniques, utilizing pixel-by-pixel spectral information as the basis for automated urban land cover detection. This paper adopts the multi-layer perceptron classifier suggested and implemented in [5]. The objective of this study is to analyse the performance and stability of this classifier - trained and tested for supervised classification (8 a priori given land use classes) of a Landsat-5 TM image (270 x 360 pixels) from the city of Vienna and its northern surroundings - along with varying the training data set in the single-training-site case. The performance is measured in terms of total classification, map user's and map producer's accuracies. In addition, the stability with initial parameter conditions, classification error matrices, and error curves are analysed in some detail. (authors' abstract) / Series: Discussion Papers of the Institute for Economic Geography and GIScience

O objetivo deste trabalho é o desenvolvimento de um método dedicado de classificação para placas de madeira utilizadas na fabricação de lápis, utilizando procedimentos de visão computacional. O processo aqui proposto, foi idealizado buscando uma metodologia que pudesse ser realizada com baixa complexidade computacional, ou seja, os cálculos dos algoritmos utilizando apenas operações simples - do tipo soma, subtração, multiplicação e divisão - em imagens em níveis de cinza. A intenção em utilizar apenas as operações básicas citadas, tem o objetivo de tornar o método implementável em arquiteturas com tecnologia VLSI, notadamente em Arquiteturas Sistólicas. O trabalho descreve o ciclo de produção do lápis localizando a etapa onde é proposta a metodologia de classificação das placas de madeira. Nesta etapa, há uma seqüência de procedimentos, descritos ao longo do trabalho, que compreendem a aquisição da imagem das placas, a extração de características das imagens, o processamento dessas características e por fim os algoritmos de classificação. Na etapa de extração de características, buscou-se com a aplicação de um método de Limiar automático que utiliza a entropia de Shannon, extrair informações suficientes para classificar adequadamente as placas de madeiras em diferentes classes, fornecendo dessa forma, um sistema ágil, repetitivo e de baixo custo para aproveitamento da madeira em diferentes produtos finais. / The objective of this work was to develop a dedicated computer vision method for the classification of wooden plates used in pencil manufacturing. The process here proposed was idealized looking for a low computational complexity methodology that could be accomplished in VLSI, as for instance using Systolic Computer Architectures made of logic arrays. The pencil cycle of production is described, locating the stage where the proposed classification methodology should be used. There is a sequence of procedures, along the work, that describe the acquisition, extraction of the characteristics and the processing of the images, and finally the classification algorithms. For the extraction of characteristics of the images, it was used an automatic method for the threshold determination, based on Shannon\'s entropy. The information supplied by the threshold determination method allows classifying the plates in different classes. The analysis of the results showed that the method performs well is repetitive and efficient on the classification and its use can be extended to classifying other final products.

[pt] O foco desta tese está direcionado a problemas de Classificação de Padrões. A proposta central é desenvolver e testar alguns novos algoritmos para ambientes supervisionados, utilizando um enfoque local- global. As principais contribuições são: (i) Desenvolvimento de método baseado em quantização vetorial com posterior classificação supervisionada local. O objetivo é resolver o problema de classificação estimando as probabilidades posteriores em regiões próximas à fronteira de decisão; (ii) Proposta do que denominamos Zona de Risco Generalizada, um método independente de modelo, para encontrar as observações vizinhas à fronteira de decisão; (iii) Proposta de método que denominamos Quantizador Vetorial das Fronteiras de Decisão, um método de classificação que utiliza protótipos, cujo objetivo é construir uma aproximação quantizada das regiões vizinhas à fronteira de decisão. Todos os métodos propostos foram testados em bancos de dados, alguns sintéticos e outros publicamente disponíveis. / [en] This thesis is focused on Pattern Classification problems. The objective is to develop and test new supervised algorithms with a local-global approach. The main contributions are: (i) A method based on vector quantization with posterior supervised local classification. The classification problem is solved by the estimation of the posterior probabilities near the decision boundary; (ii) Propose of what we call Zona de Risco Generalizada, an independent model method to find observations near the decision boundary; (iii) Propose of what we call Quantizador Vetorial das Fronteiras de Decisão, a classification method based on prototypes that build a quantized approximation of the decision boundary. All methods were tested in synthetics or real datasets.