Reconstrução e reconhecimento de imagens binárias utilizando o algoritmo Máquina de Boltzmann

The objective of this work is to use algorithms known as Boltzmann Machine to rebuild and classify patterns as images. This algorithm has a similar structure to that of an Artificial Neural Network but network nodes have stochastic and probabilistic decisions. This work presents the theoretical framework of the main Artificial Neural Networks, General Boltzmann Machine algorithm and a variation of this algorithm known as Restricted Boltzmann Machine. Computer simulations are performed comparing algorithms Artificial Neural Network Backpropagation with these algorithms Boltzmann General Machine and Machine Restricted Boltzmann. Through computer simulations are analyzed executions times of the different described algorithms and bit hit percentage of trained patterns that are later reconstructed. Finally, they used binary images with and without noise in training Restricted Boltzmann Machine algorithm, these images are reconstructed and classified according to the bit hit percentage in the reconstruction of the images. The Boltzmann machine algorithms were able to classify patterns trained and showed excellent results in the reconstruction of the standards code faster runtime and thus can be used in applications such as image recognition.

Morfometria dos forames mandibular, mentual e infraorbital de Lobo-guará Chrysocyon brachyurus, Illiger, 1815 aplicada a bloqueios anestésicos

CHAPTER II - The Chrysocyon brachyurus is the biggest South American canid which has a high frequency of dental injuries, both in wild and captivity. Thus, veterinary procedures are necessary to preserve the feeding capacity of hundreds of captive specimens worldwide. The aim of this study was to study the mandibular morphometry of maned wolf with emphasis on the establishment of anatomic references to the anesthetic blockage of the inferior alveolar and mental nerves. Therefore, 15 measurements in 22 hemimandibles of C. brachyurus adults were taken. For extra-oral technique of blockage of the inferior alveolar nerve at the level of the mandibular foramen, it is stated that the needle should be advanced, close to the medial surface of the mandibular ramus, by 11.4 mm perpendiculary from the palpable concavity. Alternatively, the needle can be introduced for 30.4 mm from the angular process at 20-25 degrees angle with the ventral margin. For blocking only the mentual nerve, it is recommended the introduction of needle for 10 mm, close to the lateral aspect of the mandibular body, at the level of the lower first premolar. The mandibular foramen showed similars position, size and symmetry in the maned wolfs specimens examined. The comparison of the data with those available for other carnivores reflects the necessity for determining these anatomical references specifically for each species.

Algorithms for super-resolution of images based on Sparse Representation and Manifolds

lmage super-resolution is defined as a class of techniques that enhance the spatial resolution of images. Super-resolution methods can be subdivided in single and multi image methods. This thesis focuses on developing algorithms based on mathematical theories for single image super­ resolution problems. lndeed, in arder to estimate an output image, we adopta mixed approach: i.e., we use both a dictionary of patches with sparsity constraints (typical of learning-based methods) and regularization terms (typical of reconstruction-based methods). Although the existing methods already per- form well, they do not take into account the geometry of the data to: regularize the solution, cluster data samples (samples are often clustered using algorithms with the Euclidean distance as a dissimilarity metric), learn dictionaries (they are often learned using PCA or K-SVD). Thus, state-of-the-art methods still suffer from shortcomings. In this work, we proposed three new methods to overcome these deficiencies. First, we developed SE-ASDS (a structure tensor based regularization term) in arder to improve the sharpness of edges. SE-ASDS achieves much better results than many state-of-the- art algorithms. Then, we proposed AGNN and GOC algorithms for determining a local subset of training samples from which a good local model can be computed for recon- structing a given input test sample, where we take into account the underlying geometry of the data. AGNN and GOC methods outperform spectral clustering, soft clustering, and geodesic distance based subset selection in most settings. Next, we proposed aSOB strategy which takes into account the geometry of the data and the dictionary size. The aSOB strategy outperforms both PCA and PGA methods. Finally, we combine all our methods in a unique algorithm, named G2SR. Our proposed G2SR algorithm shows better visual and quantitative results when compared to the results of state-of-the-art methods.

A pedagogia industrial da FIEMG: um estudo sobre o pensamento empresarial a partir da revista vida industrial (1961-1974)

The present research results of the studies and debates inherent in PhD in Education at the Graduate Program in Education at the Federal University of Uberlândia, belonging to the Research Line "Politics and Knowledge in Education." This thesis aims to analyze the contradictory meaning of education as training human history, specifically educational representative under the logic of the industrial business associated FIEMG (Federation of Industries of the State of Minas Gerais) in the context 1961-1974. This definition is justified by the historical fact that it was a period marked by the cyclical crises of capital and their impacts in the final phase of the industrialization process in Brazil: it starts with one of the apexes of economic growth in the country , driven by national developmental , continues with a severe political crisis in 1966 , also impacting the economic sphere, and finally, with the constant quest for economic stability even under high prices, hatch the factors that led to the Brazilian economy to the context of the "Economic Miracle". For this, it was necessary to link the debate between education and work from the perspective of historical materialism and dialectical their subsidies theoretical-methodological and epistemological. In the first chapter, we designed a "State of the art" category "human formation", conceived as a process of education and history, from the Marxist assumptions, aiming at the reconstruction of concepts and meanings of which is the formation of workers in contradictory logic, the bias of comprehensive training and the prospect of capital accumulation. Then, in the second chapter, we present a review about industrialization, the industrial business and its development perspective 1961-1974. The third chapter was established on a contextualization of the state and its peculiarities, the industrial business and its proposed development both nationally and at the state (Minas Gerais). Finally, in the fourth chapter, was organized dialogue with the sources, from a historical survey of the shares of the industrial business with an emphasis on education, which converged in a pedagogy industrial consolidation in line with the political and economic conditions specific period from 1961 to 1974. It has mailing bibliographic reference business thinking expressed in the concreteness of training workers and industry of Minas Gerais, in agreement with the demands of work and training of mining companies. The thesis of this study is the defense that the corporate actions which constituted pedagogy industrial concepts were articulated to political and economic development in Brazil, since the discipline to work imposed by such conceptions met the human worker training geared to the accumulation the general capital and industrial capital in particular. Establish, therefore, different logics, the state level, the scope of private foreign capital and domestic private capital, which came up the process of capital accumulation, loading, contradictorily, the possibilities of building the human beyond capital, or a teaching job.