Submitted by Automa??o e Estat?stica (sst@bczm.ufrn.br) on 2017-12-01T23:34:00Z No. of bitstreams: 1 DanielHolandaNoronha_DISSERT.pdf: 2617561 bytes, checksum: 88cfc246d074eabfd971d5b81edbf109 (MD5) / Approved for entry into archive by Arlan Eloi Leite Silva (eloihistoriador@yahoo.com.br) on 2017-12-05T21:07:17Z (GMT) No. of bitstreams: 1 DanielHolandaNoronha_DISSERT.pdf: 2617561 bytes, checksum: 88cfc246d074eabfd971d5b81edbf109 (MD5) / Made available in DSpace on 2017-12-05T21:07:18Z (GMT). No. of bitstreams: 1 DanielHolandaNoronha_DISSERT.pdf: 2617561 bytes, checksum: 88cfc246d074eabfd971d5b81edbf109 (MD5) Previous issue date: 2017-11-20 / A import?ncia do uso de FPGAs como aceleradores vem crescendo fortemente nos ?ltimos anos. Companhias como Amazon e Microsoft est?o incorporando FPGAs em seus data centers, objetivando especialmente acelerar algoritmos em suas ferramentas de busca. No centro dessas aplica??es est?o algoritmos de aprendizado de m?quina, como ? o caso da M?quina de Vetor de Suporte (SVM). Entretanto, para que essas aplica??es obtenham a acelera??o desejada, o uso eficiente dos recursos das FPGAs ? necess?rio. O projeto possui como objetivo a implementa??o paralela em hardware tanto da fase feed-forward de uma M?quina de Vetores de Suporte (SVM) quanto de sua fase de treinamento. A fase feed-forward (infer?ncia) ? implementada utilizando o kernel polinomial e de maneira totalmente paralela, visando obter a m?xima acelera??o poss?vel ao custo de uma maior utiliza??o da ?rea dispon?vel. Al?m disso, a implementa??o proposta para a infer?ncia ? capaz de computar tanto a classifica??o quanto a regress?o utilizando o mesmo hardware. J? o treinamento ? feito utilizando Otimiza??o Sequencial M?nima (SMO), possibilitando a resolu??o da complexa otimiza??o da SVM atrav?s de passos simples. A implementa??o da SMO tamb?m ? feita de modo extremamente paralelo, fazendo uso de t?cnicas para acelera??o como a cache do erro. Ademais, o Kernel Amig?vel ao Hardware (HFK) ? utilizado para diminuir a ?rea utilizada pelo kernel, permitindo que um n?mero maior de kernels seja implementado em um chip de mesmo tamanho, acelerando o treinamento. Ap?s a implementa??o paralela em hardware, a SVM ? validada por simula??o e s?o feitas an?lises associadas ao desempenho temporal da estrutura proposta, assim como an?lises associadas ao uso de ?rea da FPGA. / The importance of Field-Programmable Gate Arrays as compute accelerators has dramatically increased during the last couple of yers. Many companies such as Amazon, IBM and Microsoft included FPGAs in their data centers aiming to accelerate their search engines. In the center of those applications are many machine learning algorithms, such as Support Vector Machines (SVMs). For FPGAs to thrive in this new role, the effective usage of FPGA resources is required. The project?s main goal is the parallel FPGA implementation of both the feed-forward phase of a Support Vector Machine as well as its training phase. The feed-forward phase (inference) is implemented using the polynomial kernel in a highly parallel way in order to obtain maximum throughput at the cost of some extra area. Moreover, the inference implementation is capable of computing both classification and regression using a single hardware. The training phase of the SVM is implemented using Sequential Minimal Optimization (SMO), which enables the resolution of a complex convex optimization problem using simple steps. The SMO implementation is also highly parallel and uses some acceleration techniques, such as the error cache. Moreover, the Hardware Friendly Kernel (HFK) is used in order to reduce the kernel?s area, enabling the increase in the number of kernels per area. After the parallel implementation in hardware, the SVM is validated by simulation. Finally, analysis associated with the temporal performance of the proposed structure, as well as analysis associated with FPGA?s area usage are performed.

Submitted by Automa??o e Estat?stica (sst@bczm.ufrn.br) on 2016-07-22T15:28:33Z No. of bitstreams: 1 RamonAugustoSousaLins_DISSERT.pdf: 3798892 bytes, checksum: 0a5d06ff47c763499839e741148b3dcb (MD5) / Approved for entry into archive by Arlan Eloi Leite Silva (eloihistoriador@yahoo.com.br) on 2016-07-27T22:14:28Z (GMT) No. of bitstreams: 1 RamonAugustoSousaLins_DISSERT.pdf: 3798892 bytes, checksum: 0a5d06ff47c763499839e741148b3dcb (MD5) / Made available in DSpace on 2016-07-27T22:14:28Z (GMT). No. of bitstreams: 1 RamonAugustoSousaLins_DISSERT.pdf: 3798892 bytes, checksum: 0a5d06ff47c763499839e741148b3dcb (MD5) Previous issue date: 2015-12-04 / Conselho Nacional de Desenvolvimento Cient?fico e Tecnol?gico (CNPq) / Neste trabalho ? proposto o desenvolvimento de um sistema inteligente capaz de segmentar, contar e classificar individualmente dentes a partir de imagens fotogr?ficas digitais intraorais oclusais.O sistema proposto faz uso combinado das t?cnicas de aprendizagem de m?quina no caso a m?quina de vetor de suporte e processamento digital de imagens. Primeiramente ? feita uma segmenta??o baseada nas cores dos dentes e n?o dentes presentes na imagem atrav?s do uso de m?quina de vetores de suporte. A partir da identifica??o das regi?es de interesse, dentes e n?o dentes, os dados s?o representados de modo que a contagem, detec??o de fronteiras e classifica??o dos dentes possa ser feita. Para contagem e detec??o de fronteiras s?o utilizadas t?cnicas baseadas em operadores morfol?gicos, eros?o e transformada watershed, respectivamente. A classifica??o quanto aos tipos de dentes ? baseada na utiliza??o dos descritores de posi??o e forma, sendo esse ?ltimo definido por descritores de Fourier. O sistema portanto ? capaz de realizar a segmenta??o, a contagem e a classifica??o de dentes presentes nas imagens. / Several are the areas in which digital images are used in solving day-to-day problems. In medicine the use of computer systems have improved the diagnosis and medical interpretations. In dentistry it?s not different, increasingly procedures assisted by computers have support dentists in their tasks. Set in this context, an area of dentistry known as public oral health is responsible for diagnosis and oral health treatment of a population. To this end, oral visual inspections are held in order to obtain oral health status information of a given population. From this collection of information, also known as epidemiological survey, the dentist can plan and evaluate taken actions for the different problems identified. This procedure has limiting factors, such as a limited number of qualified professionals to perform these tasks, different diagnoses interpretations among other factors. Given this context came the ideia of using intelligent systems techniques in supporting carrying out these tasks. Thus, it was proposed in this paper the development of an intelligent system able to segment, count and classify teeth from occlusal intraoral digital photographic images. The proposed system makes combined use of machine learning techniques and digital image processing. We first carried out a color-based segmentation on regions of interest, teeth and non teeth, in the images through the use of Support Vector Machine. After identifying these regions were used techniques based on morphological operators such as erosion and transformed watershed for counting and detecting the boundaries of the teeth, respectively. With the border detection of teeth was possible to calculate the Fourier descriptors for their shape and the position descriptors. Then the teeth were classified according to their types through the use of the SVM from the method one-against-all used in multiclass problem. The multiclass classification problem has been approached in two different ways. In the first approach we have considered three class types: molar, premolar and non teeth, while the second approach were considered five class types: molar, premolar, canine, incisor and non teeth. The system presented a satisfactory performance in the segmenting, counting and classification of teeth present in the images.