The problem of estimating depth from two or more images is a fundamental problem in computer vision, which is commonly referred as to stereo matching. The applications of stereo matching range from 3D reconstruction to autonomous robot navigation. Stereo matching is particularly attractive for applications in real life because of its simplicity and low cost, especially compared to costly laser range finders/scanners, such as for the case of 3D reconstruction. However, stereo matching has its very unique problems like convergence issues in the optimisation methods, and challenges to find matches accurately due to changes in lighting conditions, occluded areas, noisy images, etc. It is precisely because of these challenges that stereo matching continues to be a very active field of research. In this thesis we develop a binocular stereo matching algorithm that works with rectified images (i.e. scan lines in two images are aligned) to find a real valued displacement (i.e. disparity) that best matches two pixels. To accomplish this our research has developed techniques to efficiently explore a 3D space, compare potential matches, and an inference algorithm to assign the optimal disparity to each pixel in the image. The proposed approach is also extended to the trinocular case. In particular, the trinocular extension deals with a binocular set of images captured at the same time and a third image displaced in time. This approach is referred as to t +1 trinocular stereo matching, and poses the challenge of recovering camera motion, which is addressed by a novel technique we call baseline recovery. We have extensively validated our binocular and trinocular algorithms using the well known KITTI and Middlebury data sets. The performance of our algorithms is consistent across different data sets, and its performance is among the top performers in the KITTI and Middlebury datasets.

It is the aim of this paper to review some of the work done on stable matching, and on stable marriage problems in particular. Variants of the stable marriage problem will be considered, and the similarities and differences from a mathematical point of view will be highlighted. The correlation between preference and stability is a main theme, and the way in which diluted or incomplete preferences affect stability is explored. Since these problems have a wide range of practical applications, it is of interest to develop useful algorithms for the derivation of solutions. Time-complexity is a key factor in designing computable algorithms, making work load a strong consideration for practical purposes. Average and worst-case complexity are discussed. The number of different solutions that are possible for a given problem instance is surprising, and counter-intuitive. This leads naturally to a study of the solution sets and the lattice structure of solutions that emerges for any stable marriage problem. Many theorems derive from the lattice structure of stable solutions and it is shown that this can lead to the design of more efficient algorithms. The research on this topic is well established, and many theorems have been proved and published, although some published proofs have omitted the detail. In this paper, the author selects some key theorems, providing detailed proofs or alternate proofs, showing the mathematical richness of this field of study. Various applications are discussed, particularly with relevance to the social sciences, although mention is made of applications in computer science, game theory, and economics. The current research that is evident in this subject area, by reference to technical papers in periodicals and on the internet, suggests that it will remain a key topic for some time to come. / MATHEMATICAL SCIENCES / MSC (MATHEMATICS)

A busca por padrões em imagens é um problema clássico em visão computacional e consiste em detectar a presença de uma dada máscara em uma imagem digital. Tal tarefa pode se tornar consideravelmente mais complexa com a invariância aos aspectos da imagem tais como rotação, escala, translação, brilho e contraste (RSTBC - rotation, scale, translation, brightness and contrast). Um algoritmo de busca de máscara foi recentemente proposto. Este algoritmo, chamado de Ciratefi, é invariante aos aspectos RSTBC e mostrou-se bastante robusto. Entretanto, a execução deste algoritmo em um computador convencional requer diversos segundos. Além disso, sua implementação na forma mais geral em hardware é difícil pois há muitos parâmetros ajustáveis. Este trabalho propõe o projeto de um software que gera automaticamente módulos compiláveis em Hardware Description Logic (VHDL) que implementam o filtro circular do algoritmo Ciratefi em dispositivos Field Programmable Gate Array (FPGA). A solução proposta acelera o tempo de processamento de 7s (em um PC de 3GHz) para 1,367ms (em um dispositivo Stratix III da Altera). Esta performance excelente (mais do que o necessário em sistemas em tempo-real) pode levar a sistemas de visão computacional de alta performance e de baixo custo. / Template matching is a classical problem in computer vision. It consists in detecting the presence of a given template in a digital image. This task becomes considerably more complex with the invariance to rotation, scale, translation, brightness and contrast (RSTBC). A novel RSTBC-invariant robust template matching algorithm named Ciratefi was recently proposed. However, its execution in a conventional computer takes several seconds. Moreover, the implementation of its general version in hardware is difficult, because there are many adjustable parameters. This work proposes a software that automatically generates compilable Hardware Description Logic (VHDL) modules that implement the circular filter of the Ciratefi template matching algorithm in Field Programmable Gate Array (FPGA) devices. The proposed solution accelerates the time to process a frame from 7s (in a 3GHz PC) to 1.367ms (in Altera Stratix III device). This excellent performance (more than the required for a real-time system) may lead to cost-effective high-performance coprocessing computer vision systems.

FundaÃÃo Cearense de Apoio ao Desenvolvimento Cientifico e TecnolÃgico / No Brasil, a busca pela reduÃÃo das ineficiÃncias observadas na alocaÃÃo de vagas em instituiÃÃes de ensino superior via o tradicional vestibular levou à formulaÃÃo e implantaÃÃo de um mecanismo alternativo de seleÃÃo para admissÃo superior: o Sistema de SeleÃÃo Unificada (SISU), criado em 2010. O mecanismo, tecnicamente falando, à um algoritmo de matching com as seguintes caracterÃsticas: i) cada estudante que recebe oferta de matrÃcula decide por aceitar ou rejeitar a oferta recebida; ii) rejeiÃÃes de ofertas provocam a realizaÃÃo de novas propostas; e iii) propostas sÃo aceitas temporariamente, podendo cada oferta aceita ser âtrocadaâ por uma oferta considerada âmelhorâ. Ou seja, o SISU à um mecanismo semelhante ao Algoritmo Deferred Acceptance (Algoritmo Gale-Shapley) com os cursos propondo. Apesar da importÃncia do SISU, a literatura econÃmica sobre o tema à basicamente inexistente. Nesse sentido, a presente dissertaÃÃo buscou, à luz da teoria dos jogos, entender e caracterizar os incentivos propiciados pelo SISU atravÃs de dois mecanismos teÃricos desenvolvidos, o SISUα e o SISUβ. Ambos sÃo modelados como mecanismos de matching dinÃmicos. Caracterizamos estratÃgias nÃo dominadas para o SISUβ e o SISUα. Utilizando o SISUα como a melhor aproximaÃÃo disponÃvel para o SISU, concluÃmos que a introduÃÃo do SISU apresentou um importante avanÃo em relaÃÃo ao vestibular em termos de ganhos de eficiÃncia do matching entre alunos e cursos. / In Brazil, the quest for reducing observed inefficiencies in the allocation of seats in higher education institutions through traditional examination (Vestibular) led to the formulation and implementation of an alternative mechanism of selection: the Unified Selection System (SISU), created in 2010. Technically speaking, the mechanism is a matching algorithm with the following characteristics: i) each student who receives an offer decides to accept or reject the received offer; ii) rejections of offers prompt new proposals; iii) proposals are accepted temporarily, and each offer can be âreplacedâ by an offer considered âbetterâ. That is, the SISU is a mechanism similar to the Deferred Acceptance Algorithm (Gale-Shapley Algorithm). Despite the importance of SISU, the economic literature on the subject is basically nonexistent. In this sense, this dissertation sought, in light of game theory, to understand and characterize the incentives provided by SISU through two developed theoretical mechanisms, the SISUα and SISUβ. Both are modeled as dynamic matching mechanisms. We characterize undominated strategies for SISUβ and SISUα. Using SISUα as the best approximation available to SISU, we conclude that the introduction of SISU showed a significant improvement on Vestibular in terms of gains of efficiency from matching between students and courses.

Neste trabalho analisou-se o impacto das remessas financeiras sobre os patrimônios dos ativos nos domicílios, no país de origem. Utilizamos dados da Encuesta Permanente de Hogares de 2008, os quais foram fornecidos pelo órgão Dirección General de Estadística, Encuestas y Censos do Paraguai. Com estes dados realizamos uma comparação entre os domicílios com e sem remessas, independente de terem ou não emigrantes. A metodologia utilizada para realizar a comparação foi o Propensity Score Matching (PSM) com dois algoritmos, Vizinhos Próximos e Kernel. Os patrimônios dos ativos são carros, aluguéis e combinados. Os resultados em todos os casos foram estatisticamente significativos, porém, negativos. Com amesma metodologia e os mesmos dados, comparamos apenas os domicílios com emigrantes, e os resultados continuaram negativos, porém, com nível de significância menor. / In this study we analyzed the impact of remittances on financial assets in the wealth of households in the country of origin. We use data from the Encuesta Permanente Hogares 2008, which were provided by the agency Dirección General de Estadística, Encuestas y Censos of Paraguay. With these data we performed a comparison between households with and without remittances, regardless of whether or not immigrants. The methodology for conducting the comparison was the Propensity Score Matching (PSM) with two algorithms, Kernel and Nearest Neighbors. The stockholders\' assets are cars, rentals and combined. The results in all cases were statistically significant, but negative. With the same methodology and same data, comparing only households with migrants, the results remained negative, but with lower significance level.

O objetivo deste estudo é analisar o impacto da saúde sobre o desempenho escolar de alunosda quarta série do ensino fundamental em todo o Brasil no ano de 2005 e 2007. No Brasil, a carência de bases de dados contendo simultaneamente indicadores de saúde e de desempenho escolar dificulta as pesquisas sobre o tema. Como alternativa, este estudo utiliza dados agregados referentes à oferta de serviços de saúde no município e microrregião, provenientes do DATASUS e da Pesquisa Assistência MédicoSanitária, como proxy da saúde dos alunos da quarta série nos municípios. Os dados para o desempenho escolar dos alunos são provenientes da Prova Brasil 2005 e 2007. Dessa forma, a unidade de observação passa a ser a microrregião, e não o aluno. A hipótese a ser testada é se a oferta de serviços de saúde na microrregião afeta o desempenho médio dos alunos dessa microrregião. Utilizase inicialmente uma análise de regressão linear contendo as variáveis indicadoras de saúde como regressores em uma crosssection para 2005. Em seguida, a análise é feita em painel, com observações de 2005 e 2007. Os resultados indicam que há impacto positivo, mas apenas em alguns indicadores e, mesmo assim, o impacto é pequeno. Outra abordagem é a utilização de surtos de dengue em municípios para determinar se existe impacto negativo no desempenho escolar dos alunos de municípios com surto. Para estimação nessa abordagem foi utilizado matching . Os resultados dessa análise indicam impacto negativo, porém sensível à escolha das variáveis e de algoritmos de estimação. / The objective of this study is to analyze the impact of health on the school proficiency of students in the fourth grade of elementary school in Brazil in the years 2005 and 2007. In Brazil, the lack of data containing simultaneously indicators of health and school proficiency makes it harder to research on the subject. As an alternative, this study utilizes aggregated data, like the offer of health services in cities, taken from DATASUS and Pesquisa Assistência MédicoSanitária, as a proxy for the health of students of the fourth grade in the cities. The data for school proficiency of students comes from Prova Brasil 2005 and 2007. That way, the observation unit becomes the city, not the student. The hypothesis to be tested is whether or not the offer of health services in a city can affect the school proficiency. Initially a linear regression is utilized containing indicators of health variables as regressors in a crosssection in 2005. Next, the analysis is made in panel data, with observations of 2005 and 2007. The results show that there is positive impact, but only in a few indicators of health and, even then, the impact is small. Another approach was to utilize dengue outbreaks in cities as a way to determine whether there was a negative impact in the school proficiency of students that resided in a city with an outbreak. Matching was used for the estimation in this approach. The results of this analysis indicate a negative impact, but sensible to the choice of variables and estimation algorithms.

A presente dissertação visa avaliar, a partir do uso dos dados da Pesquisa de Atividade Econômica Paulista - PAEP - do ano de 1996, da RAIS - Relação Anual de Informações Sociais de 1992, 1993, 1997 e 1998 e da SECEX - Secretaria do Comércio Exterior - dos anos de 1992 e 1993, a relação entre a atividade de inovação tecnológica e os potenciais impactos produzidos sobre o nível de emprego e sobre o nível de renda na indústria paulista entre os anos de 1997 e 1998. Como a inovação não é um evento aleatório, há o viés de seleção da amostra, de tal sorte que uma comparação direta entre os efeitos da inovação sobre certas variáveis daquelas firmas que inovaram com aquelas que não inovaram, na situação de ter havido inovação, produziria vieses. Tal problema é resolvido neste trabalho por meio do Propensity Score Matching, que visa o pareamento de unidades tratamento - controle para a obtenção dos resultados intentados, a saber, a variação na renda e no nível de emprego. Ademais, de modo a inibir a presença de efeitos não observáveis, recorre-se ao método de Diferenças em Diferenças (DID). Os resultados obtidos atestam para um aumento, na média, do nível de emprego para quase todos os tipos de atividade inovativa empreendidas. Em contrapartida, os achados se revelaram estatisticamente insignificantes para a variação de renda, também em quase todos os tipos de inovação. / This work intends to analyze, by the usage of three different data sources, PAEP (1996), RAIS - Relação Anual de Informações Sociais (1992, 1993, 1997, and 1998) and SECEX - Secretaria do Comércio Exterior (from 1992, and 1993) the relationship, if that exists, between innovation and its potential impacts over employment level as well as salaries during the 1997 - 1998 period. Once innovation activity cannot be considered as a random event, we incur in the well known selection bias problem. Consequently, a direct comparison of the innovation impacts between those firms which effectively innovated with those which did not innovate would be misleading. Such a problem may be corrected with the Propensity Score Matching, which consists in the establishment of treatment - control pairs in order to obtain the results previously searched. Moreover, in an attempt to control for the unobserved effects, it is applied the Differences in Differences Methodology (DID). Results obtained show that, on the one hand, for almost all the innovation activities considered, there was an increment in the level of employment. On the other hand, for most of the cases, there were not statistically significant results for an impact of innovation activities on the level of salaries.

O objetivo deste trabalho é contribuir para literatura empírica que avalia o impacto dos incentivos públicos à pesquisa e desenvolvimento (P&D) com dados de firmas brasileiras. Em particular foi avaliado o impacto da Lei do Bem, instrumento de incentivo fiscal à atividade de pesquisa e desenvolvimento privado. Essa avaliação foi conduzida a partir de estimações de modelos econométricos com microdados de empresas industriais brasileiras. Foi aplicado a técnica de matching e realizadas estimações de modelos empíricos de investimento com dados em painel. O impacto foi avaliado considerando toda amostra de empresas industriais e por intensidade tecnológica, adicionalmente foi analisado o efeito de dosagem. Os resultados trazem evidências que existe impacto positivo no dispêndio em P&D nas firmas, rejeitando a hipótese de crowding-out. / The objective of this work is to contribute to the empirical literature that evaluates the effectiveness of public support on private R&D with Brazilian firms data. In particular was evaluated an instrument of fiscal incentive called ,,Lei do Bem\". The evaluation was conducted applying econometric approach using microdata of industrial firms. A matching was conducted and estimated empirical investment equation with panel data. The effect was evaluated in full sample and dividing by technological intensity, the dosage effect was also considered. The results indicate a positive impact on the expenditure in R&D, rejecting the crowding-out hypothesis.

O presente estudo investigou comportamentos simbólicos nos portadores de Síndrome de Apert por meio do paradigma da equivalência de estímulos, uma vez que este pode fornecer subsídios para a compreensão do comportamento humano complexo, tais como o comportamento simbólico e a linguagem. Foram propostos três experimentos cujos participantes, com idades entre 16 e 21 anos, foram submetidos aos Experimentos 1 e 2 (P1, P2 e P3) e submetidos ao Experimento 3 (P4 e P5). No Experimento 1, o objetivo foi verificar a formação de classes de estímulos equivalentes após o treino de relações condicionais, por meio de uma tarefa de matching to sample (MTS) com atraso de 0 s e três estímulos de comparação. Após o participante atingir um percentual de acerto de 94% no bloco, foram conduzidos testes de equivalência (CA), simetria (BA e CB) e transitividade (AC). De maneira geral, houve falha nos testes, com relativa preservação do teste de simetria. No Experimento 2, os participantes foram expostos à tarefa de MTS com atraso de 0 s e três estímulos de comparação, mas estes e o estímulo modelo apresentavam-se inicialmente cobertos, tornando-se visíveis somente após a emissão de respostas de observação (ROs). Foram introduzidas ROs na tarefa de MTS, com o objetivo de identificar que tipos de controle (seleção ou rejeição) poderiam estar em operação na formação de classes de estímulos equivalentes. Nos testes, o participante P3 formou classes de equivalência, contudo, os participantes P1 e P2 continuaram a apresentar falha na formação de equivalência, com preservação do teste de simetria. No Experimento 3, foram apresentadas variações metodológicas com o objetivo de favorecer o estabelecimento de um ou de outro tipo de controle (seleção ou rejeição), durante o treino das relações condicionais e verificar os efeitos destas variações no desempenho obtido nos testes. O participante P4 foi submetido primeiramente à situação na qual a terceira RO produzia o estímulo correto em 80% das tentativas (80%/3ª S+). Posteriormente 9 à realização das tarefas sob esta condição, foi submetido à situação na qual a primeira RO produzia o estímulo correto em 80% das tentativas (80%/1ª S+). O participante P5 foi exposto à ordem inversa: primeiramente à situação 80%/1ª S+ e posteriormente à situação 80%/3ª S+. Nos testes, em ambas as situações, o participante P4 foi capaz de aprender as relações de equivalência propostas pelo experimentador. Diferentemente, o responder do participante P5 não se apresentou sob controle discriminativo das contingências programadas pelo experimentador. De um modo geral, os resultados obtidos pelo presente estudo acrescidos das análises do teste PEABODY, sugerem que as dificuldades para a formação de classes de equivalência em participantes com baixo funcionamento lingüístico podem decorrer, em grande parte, de preparação e adaptação insuficiente de procedimentos para esse tipo de população / The present study investigated the symbolic behaviors of patients with Apert Syndrome through the paradigm of stimulus equivalence, since it may provide a basis for understanding complex human behavior, such as symbolic behavior and language. It had been proposed three experiments which participants, aged between 16 and 21 years, were submitted to Experiments 1 and 2 (P1, P2 and P3) and to Experiment 3 (P4 e P5). In Experiment 1, the objective was to assess the formation of classes of equivalent stimuli after training of conditional relations, through a task of matching to sample (MTS) and delay of 0s and three comparison stimuli. After the participant reaches a 94% criterion for accuracy in the block, tests were conducted for equivalence (CA), symmetry (BA and CB) and transitivity (AC). In general, the participants failed the tests, with relative preservation of the symmetry test. In Experiment 2, the participants were exposed to the task of MTS with delay of 0s and three comparison stimuli, but these and the sample stimulus were initially covered, becoming visible only after the observation responses (OR´s) issuance. ORs were introduced in the MTS task, with the objective to assess what kind of different controls (selection and rejection) could be involved in the formation of classes of equivalent stimuli. In tests, participant P3 showed the formation of classes of equivalent stimuli; however, participants P1 and P2 had failed to form equivalence, preserving the symmetry test. In Experiment 3 were presented methodological variations with the objective of encouraging the establishment of one or another type of control (selection or rejection) during the training of conditional relations and verify the effects of these variations in performance obtained in tests. Participant P4 was, at first, subjected to the situation which the third OR produced the correct stimulus in 80% of attempts (80% / 3rd S +). After the tasks under this condition were performed, P4 underwent a situation which the first OR produced the correct stimulus in 80% of attempts (80% / 1st S +). The participant P5 was exposed to the reverse order: first of all to the situation 80% / 1st S + and subsequently to the situation 80% / 3rd S +. In the tests, in both situations, the participant P4 was able to learn the equivalence 11 relations proposed by the experimenter. In contrast, the response of the participant P5 was not under discriminative control of the contingencies programmed by the experimenter. In general, the results obtained by this study plus the analysis of the PEABODY data suggests that difficulties in the formation of equivalence classes in participants with low language functioning may be due, in large part from inadequate preparation and adjustment procedures for this type of population

Context: Determining attribute correspondence is the most important, time consuming and knowledge intensive part during databases integration. It is also used in other data manipulation applications such as data warehousing, data design, semantic web and e-commerce. Objectives: In this thesis the aim is to investigate how to find the types of correspondence between the attributes of heterogeneous datasets when schema design information of the data sets is unknown. Methods: A literature review was conducted to extract the knowledge related to the approaches that are used to find the correspondence between the attributes of heterogeneous datasets. Extracted knowledge from the literature review is used in developing an instance based approach for finding types of correspondence between the attributes of heterogeneous datasets when schema design information is unknown. To validate the proposed approach an experiment was conducted in the real environment using the data provided by the Telecom Industry (Ericsson) Karlskrona. Evaluation of the results was carried using the well known and mostly used measures from information retrieval field precision, recall and F-measure. Results: To find the types of correspondence between the attributes of heterogeneous datasets, good results depend on the ability of the algorithm to avoid the unmatched pairs of rows during the Row Similarity Phase. An evaluation of proposed approach is performed via experiments. We found 96.7% (average of three experiments) F-measure. Conclusions: The analysis showed that the proposed approach was feasible to be used and it provided users a mean to find the corresponding attributes and the types of correspondence between corresponding attributes, based on the information extracted from the similar pairs of rows from the heterogeneous data sets where their similarity based on the same common primary keys values.