This thesis includes three separate research projects focusing on computer vision principles and deep learning pattern recognition problems. Chapter 3 entails color quantization applications using traditional Kmeans clustering techniques and random selection of color techniques within the red, green, blue (RGB) color space to maintain a high-quality image while significantly reducing image file size. Chapter 4 consists of a handwriting character recognition algorithm using backpropagation to classify 70,000 handwritten values from US Census Bureau employees and high school students. Chapter 5 proposes a novel classification technique for 109,446 unique heartbeat samples to identify areas of interest and assist medical professionals in diagnosing heart problems.

The self-organizing map (SOM) is an unsupervised learning algorithm which has been successfully applied to various applications. One of advantages of SOM is it maintains an incremental property to handle data on the fly. In the last several decades, there have been variants of SOM used in many application domains. In this dissertation, two new SOM algorithms are developed for image quantization and compression. The first algorithm is a sample-size adaptive SOM algorithm that can be used for color quantization of images to adapt to the variations of network parameters and training sample size. The sweep size of neighborhood function is modulated by the size of the training data. In addition, the minimax distortion principle which is modulated by training sample size is used to search the winning neuron. Based on the sample-size adaptive self-organizing map, we use the sampling ratio of training data, rather than the conventional weight change between adjacent sweeps, as a stop criterion. As a result, it can significantly speed up the learning process. Experimental results show that the proposed sample-size adaptive SOM achieves much better PSNR quality, and smaller PSNR variation under various combinations of network parameters and image size. The second algorithm is a novel classified SOM method for edge preserving quantization of images using an adaptive subcodebook and weighted learning rate. The subcodebook sizes of two classes are automatically adjusted in training iterations based on modified partial distortions that can be estimated incrementally. The proposed weighted learning rate updates the neuron efficiently no matter of how large the weighting factor is. Experimental results show that the proposed classified SOM method achieves better quality of reconstructed edge blocks and more spread out codebook and incurs a significantly less computational cost as compared to the competing methods.

In this thesis, we introduce a radial basis function network approach to solve a color image classification problem in a real time industrial application. Radial basis function networks are employed to classify the images of finished wooden parts in terms of their color and species. Other classification methods are also examined in this work. The minimum distance classifiers are presented since they have been employed by the previous research. We give brief definitions about color space, color texture, color quantization, color classification methods. We also give an intensive review of radial basis functions, regularization theory, regularized radial basis function networks, and generalized radial basis function networks. The centers of the radial basis functions are calculated by the k-means clustering algorithm. We examine the k-means algorithm in terms of starting criteria, the movement rule, and the updating rule. The dilations of the radial basis functions are calculated using a statistical method. Learning classifier systems are also employed to solve the same classification problem. Learning classifier systems learn the training samples completely whereas they are not successful to classify the test samples. Finally, we present some simulation results for both radial basis function network method and learning classifier systems method. A comparison is given between the results of each method. The results show that the best classification method examined in this work is the radial basis function network method. / Master of Science

Inspirado nas técnicas utilizadas por pintores que sobrepõem camadas de tintas de diversos matizes na geração de uma tela artística e também observando-se a distribuição da quantidade dos cones na retina do olho humano na interpretação destas cores, este trabalho propõe uma técnica de processamento de imagens baseada na mistura de cores. Trata-se de um método de quantização de cores estático que expressa a proporção das cores preto, azul, verde, ciano, vermelho, magenta, amarelo e branco obtida pela representação binária da cor que compõe os pixels de uma imagem RGB com 8 bits por canal. O histograma da mistura é denominado de misturograma e gera planos que interceptam o espaço RGB, definindo o espaço de cor HSM (Hue, Saturation and Mixture). A posição destes planos dentro do cubo RGB é modelada por meio da distribuição dos cones sensíveis aos comprimentos de onda curta (Short), média (Middle) e longa (Long) consideradas para a retina humana. Para demonstrar a aplicabilidade do espaço de cor HSM, é proposta, neste trabalho, a segmentação dos pixels de uma imagem digital em pele humana ou não pele com o uso dessa nova abordagem. Para análise de desempenho da mistura de cores foi implementado um método tradicional no espaço de cor RGB e também usando uma distribuição Gaussiana nos espaços de cores HSV e HSM. Os resultados obtidos demonstram o potencial da técnica que emprega a mistura de cores para a segmentação de imagens digitais coloridas. Verificou-se também que, baseando-se apenas na camada mais significativa da mistura de cores, gera-se a imagem esboço de uma imagem facial denominada esboço da face. Os resultados obtidos comprovam o bom desempenho do esboço da face em aplicações CBIR. / Inspired on the techniques used by painters to overlap layers of various hues of paint to create oil paintings, and also on observations of the distribution of cones in human retina for the interpretation of these colors, this thesis proposes an image processing technique based on color mixing. This is a static color quantization method that expresses the mixture of black, blue, green, cyan, red, magenta, yellow and white colors quantified by the binary weight of the color that makes up the pixels of an RGB image with 8 bits per channel. The mixture histogram, called a mixturegram, generates planes that intersect the RGB color space, defining the HSM (Hue, Saturation and Mixture) color space. The position of these planes inside the RGB cube is modeled by the distribution of cones sensitive to the short (S), middle (M) and long (L) wave lengths of the human retina. To demonstrate the applicability of the HSM color space, this thesis proposes the segmentation of the pixels of a digital image of human skin or non-skin using this new approach. The performance of the color mixture is analyzed by implementing a traditional method in the RGB color space and by a Gaussian distribution in the HSV and HSM color spaces. The results demonstrate the potential of the proposed technique for color image segmentation. It was also noted that, based only on the most significant layer of the colors mixture, it is possible generates the face sketch image. The results show the performance of the face sketch image in CBIR applications.

Inspirado nas técnicas utilizadas por pintores que sobrepõem camadas de tintas de diversos matizes na geração de uma tela artística e também observando-se a distribuição da quantidade dos cones na retina do olho humano na interpretação destas cores, este trabalho propõe uma técnica de processamento de imagens baseada na mistura de cores. Trata-se de um método de quantização de cores estático que expressa a proporção das cores preto, azul, verde, ciano, vermelho, magenta, amarelo e branco obtida pela representação binária da cor que compõe os pixels de uma imagem RGB com 8 bits por canal. O histograma da mistura é denominado de misturograma e gera planos que interceptam o espaço RGB, definindo o espaço de cor HSM (Hue, Saturation and Mixture). A posição destes planos dentro do cubo RGB é modelada por meio da distribuição dos cones sensíveis aos comprimentos de onda curta (Short), média (Middle) e longa (Long) consideradas para a retina humana. Para demonstrar a aplicabilidade do espaço de cor HSM, é proposta, neste trabalho, a segmentação dos pixels de uma imagem digital em pele humana ou não pele com o uso dessa nova abordagem. Para análise de desempenho da mistura de cores foi implementado um método tradicional no espaço de cor RGB e também usando uma distribuição Gaussiana nos espaços de cores HSV e HSM. Os resultados obtidos demonstram o potencial da técnica que emprega a mistura de cores para a segmentação de imagens digitais coloridas. Verificou-se também que, baseando-se apenas na camada mais significativa da mistura de cores, gera-se a imagem esboço de uma imagem facial denominada esboço da face. Os resultados obtidos comprovam o bom desempenho do esboço da face em aplicações CBIR. / Inspired on the techniques used by painters to overlap layers of various hues of paint to create oil paintings, and also on observations of the distribution of cones in human retina for the interpretation of these colors, this thesis proposes an image processing technique based on color mixing. This is a static color quantization method that expresses the mixture of black, blue, green, cyan, red, magenta, yellow and white colors quantified by the binary weight of the color that makes up the pixels of an RGB image with 8 bits per channel. The mixture histogram, called a mixturegram, generates planes that intersect the RGB color space, defining the HSM (Hue, Saturation and Mixture) color space. The position of these planes inside the RGB cube is modeled by the distribution of cones sensitive to the short (S), middle (M) and long (L) wave lengths of the human retina. To demonstrate the applicability of the HSM color space, this thesis proposes the segmentation of the pixels of a digital image of human skin or non-skin using this new approach. The performance of the color mixture is analyzed by implementing a traditional method in the RGB color space and by a Gaussian distribution in the HSV and HSM color spaces. The results demonstrate the potential of the proposed technique for color image segmentation. It was also noted that, based only on the most significant layer of the colors mixture, it is possible generates the face sketch image. The results show the performance of the face sketch image in CBIR applications.

Les entreprises doivent traiter quotidiennement de gros volumes de documents papiers de toutes sortes. Automatisation, traçabilité, alimentation de systèmes d’informations, réduction des coûts et des délais de traitement, la dématérialisation a un impact économique évident. Pour respecter les contraintes industrielles, les processus historiques d’analyse simplifient les images grâce à une séparation fond/premier-plan. Cependant, cette binarisation peut être source d’erreurs lors des étapes de segmentation et de reconnaissance. Avec l’amélioration des techniques, la communauté d’analyse de documents a montré un intérêt croissant pour l’intégration d’informations colorimétriques dans les traitements, ceci afin d’améliorer leurs performances. Pour respecter le cadre imposé par notre partenaire privé, l’objectif était de mettre en place des processus non supervisés. Notre but est d’être capable d’analyser des documents même rencontrés pour la première fois quels que soient leurs contenus, leurs structures, et leurs caractéristiques en termes de couleurs. Les problématiques de ces travaux ont été d’une part l’identification d’un nombre raisonnable de couleurs principales sur une image ; et d’autre part, le regroupement en couches couleur cohérentes des pixels ayant à la fois une apparence colorimétrique très proche, et présentant une unité logique ou sémantique. Fournies sous forme d’un ensemble d’images binaires, ces couches peuvent être réinjectées dans la chaîne de dématérialisation en fournissant une alternative à l’étape de binarisation classique. Elles apportent en plus des informations complémentaires qui peuvent être exploitées dans un but de segmentation, de localisation, ou de description. Pour cela, nous avons proposé une segmentation spatio-colorimétrique qui permet d’obtenir un ensemble de régions locales perceptuellement cohérentes appelées superpixels, et dont la taille s’adapte au contenu spécifique des images de documents. Ces régions sont ensuite regroupées en couches couleur globales grâce à une analyse multi-résolution. / Industrial companies receive huge volumes of documents everyday. Automation, traceability, feeding information systems, reducing costs and processing times, dematerialization has a clear economic impact. In order to respect the industrial constraints, the traditional digitization process simplifies the images by performing a background/foreground separation. However, this binarization can lead to some segmentation and recognition errors. With the improvements of technology, the community of document analysis has shown a growing interest in the integration of color information in the process to enhance its performance. In order to work within the scope provided by our industrial partner in the digitization flow, an unsupervised segmentation approach was chosen. Our goal is to be able to cope with document images, even when they are encountered for the first time, regardless their content, their structure, and their color properties. To this end, the first issue of this project was to identify a reasonable number of main colors which are observable on an image. Then, we aim to group pixels having both close color properties and a logical or semantic unit into consistent color layers. Thus, provided as a set of binary images, these layers may be reinjected into the digitization chain as an alternative to the conventional binarization. Moreover, they also provide extra-information about colors which could be exploited for segmentation purpose, elements spotting, or as a descriptor. Therefore, we have proposed a spatio-colorimetric approach which gives a set of local regions, known as superpixels, which are perceptually meaningful. Their size is adapted to the content of the document images. These regions are then merged into global color layers by means of a multiresolution analysis.