abstract: Detection of extruded features like rooftops and trees in aerial images automatically is a very active area of research. Elevated features identified from aerial imagery have potential applications in urban planning, identifying cover in military training or flight training. Detection of such features using commonly available geospatial data like orthographic aerial imagery is very challenging because rooftop and tree textures are often camouflaged by similar looking features like roads, ground and grass. So, additonal data such as LIDAR, multispectral imagery and multiple viewpoints are exploited for more accurate detection. However, such data is often not available, or may be improperly registered or inacurate. In this thesis, we discuss a novel framework that only uses orthographic images for detection and modeling of rooftops. A segmentation scheme that initializes by assigning either foreground (rooftop) or background labels to certain pixels in the image based on shadows is proposed. Then it employs grabcut to assign one of those two labels to the rest of the pixels based on initial labeling. Parametric model fitting is performed on the segmented results in order to create a 3D scene and to facilitate roof-shape and height estimation. The framework can also benefit from additional geospatial data such as streetmaps and LIDAR, if available. / Dissertation/Thesis / M.S. Computer Science 2013

abstract: Video object segmentation (VOS) is an important task in computer vision with a lot of applications, e.g., video editing, object tracking, and object based encoding. Different from image object segmentation, video object segmentation must consider both spatial and temporal coherence for the object. Despite extensive previous work, the problem is still challenging. Usually, foreground object in the video draws more attention from humans, i.e. it is salient. In this thesis we tackle the problem from the aspect of saliency, where saliency means a certain subset of visual information selected by a visual system (human or machine). We present a novel unsupervised method for video object segmentation that considers both low level vision cues and high level motion cues. In our model, video object segmentation can be formulated as a unified energy minimization problem and solved in polynomial time by employing the min-cut algorithm. Specifically, our energy function comprises the unary term and pair-wise interaction energy term respectively, where unary term measures region saliency and interaction term smooths the mutual effects between object saliency and motion saliency. Object saliency is computed in spatial domain from each discrete frame using multi-scale context features, e.g., color histogram, gradient, and graph based manifold ranking. Meanwhile, motion saliency is calculated in temporal domain by extracting phase information of the video. In the experimental section of this thesis, our proposed method has been evaluated on several benchmark datasets. In MSRA 1000 dataset the result demonstrates that our spatial object saliency detection is superior to the state-of-art methods. Moreover, our temporal motion saliency detector can achieve better performance than existing motion detection approaches in UCF sports action analysis dataset and Weizmann dataset respectively. Finally, we show the attractive empirical result and quantitative evaluation of our approach on two benchmark video object segmentation datasets. / Dissertation/Thesis / M.S. Computer Science 2013

abstract: Two critical limitations for hyperspatial imagery are higher imagery variances and large data sizes. Although object-based analyses with a multi-scale framework for diverse object sizes are the solution, more data sources and large amounts of testing at high costs are required. In this study, I used tree density segmentation as the key element of a three-level hierarchical vegetation framework for reducing those costs, and a three-step procedure was used to evaluate its effects. A two-step procedure, which involved environmental stratifications and the random walker algorithm, was used for tree density segmentation. I determined whether variation in tone and texture could be reduced within environmental strata, and whether tree density segmentations could be labeled by species associations. At the final level, two tree density segmentations were partitioned into smaller subsets using eCognition in order to label individual species or tree stands in two test areas of two tree densities, and the Z values of Moran's I were used to evaluate whether imagery objects have different mean values from near segmentations as a measure of segmentation accuracy. The two-step procedure was able to delineating tree density segments and label species types robustly, compared to previous hierarchical frameworks. However, eCognition was not able to produce detailed, reasonable image objects with optimal scale parameters for species labeling. This hierarchical vegetation framework is applicable for fine-scale, time-series vegetation mapping to develop baseline data for evaluating climate change impacts on vegetation at low cost using widely available data and a personal laptop. / Dissertation/Thesis / M.A. Geography 2013

El desarrollo de técnicas matemáticas para mejorar la visualización de imágenes médicas DICOM, mediante segmentación y procesamiento morfológico es el principal objetivo del presente proyecto de tesis. Para lograr este fin, se recopilaron imágenes médicas de un Tomógrafo Computarizado en diferentes estudios especializados de UROTEM, Tórax y de Cerebro, los cuales fueron procesados mediante algoritmos matemáticos desarrollados en el software Matlab. En el primer capítulo se plantea el problema; la necesidad de un post procesamiento de imágenes médicas, se fundamenta la importancia de esta investigación y se identifican los objetivos a lograr, así como también, se describen los antecedentes de esta investigación. Como marco teórico, en el segundo capítulo, se explica brevemente el funcionamiento de un Tomógrafo Computarizado, la adquisición de una imagen y el desarrollo del formato DICOM como un estándar internacional para imágenes médicas. Las técnicas matemáticas se describen en el capítulo tres; segmentación de imágenes, mediante el método de N. Otsu y procesamiento morfológico, detallándose su formulación y se definen las funciones empleadas en el software Matlab basadas en estas técnicas. El desarrollo de técnicas computacionales son descritas en el capítulo cuatro, donde se presenta por cada estudio; un diagrama de bloques y el desarrollo del procedimiento del algoritmo en Matlab, Finalmente, en el capítulo cinco, se muestran las pruebas y resultados de las imágenes DICOM de cinco pacientes por estudio, identificándose las patologías de forma automática. The mathematical technics to improve the visualization of medical images DICOM, through segmentation and morphological processing is the main goal of the following thesis project. In order to achieve this purpose, we collected medical images from a computed tomography scanner in three different exams, UROTEM, Thorax, and Brain. These images were processed through the mathematical algorithms developed in the software Matlab. The first chapter presents the problem; the necessity to post processing the medical images, the importance of the investigation as a tool to help in the medical diagnosis and the identification of a pathology automatically, the identification of the objectives to accomplish, and also it describes the antecedents of medical images processing developed from different authors. As a theory, in the second chapter, explain briefly the working basis of the Computed tomography scanner, the acquisition of an image and the developing of DICOM format as an international standard for medical images. The mathematical technics, are described in the chapter three; segmentation of the images, through the Otsu method and morphological processing, explaining its formula and function in Matlab Software. The developing of the computational technics are described in chapter four , UROTEM, Thorax and Brain; a flow chart of each exam and the step by step procedure of the Matlab algorithm, which applied the segmentation and morphological processing in the images.

La classification d'une image ETM+ de Landsat a été réalisée pour la cartographie des types de couverture du sol dans la moitié sud du parc national du Canada Auyuittuq. Le projet fait appel à l'analyse d'image par objets spatiaux (object-based image analysis ). Le logiciel eCognition 4.0 permet une segmentation hiérarchique de l'image qui est analogue au concept de l'inventaire écologique des parcs nationaux. Un territoire d'environ 8 300 km 2 a été cartographié à trois niveaux de perception différents à partir d'une image acquise le 13 août 2000. Un modèle numérique d'altitude fut incorporé au projet et de nombreux indices spectraux ont été calculés à partir des données ETM + . Le niveau de segmentation brute comporte 375 312 objets regroupés en 36 classes. À ce niveau, la structure de classification repose sur 118 règles référant aux paramètres spectraux, spatiaux et topographiques des segments. Ces règles combinent des systèmes de seuillages chiffrés et des opérations de tri au plus proche voisin. L'attribution des segments aux classes du projet est tributaire de ces règles et répond à une logique floue. À la suite d'une fusion de segments et d'un premier regroupement de classes, on obtient le second niveau du projet, qui compte 102 239 objets et 28 classes thématiques. Ce niveau s'apparente aux"écotypes" au sens de l'inventaire écologique des parcs nationaux. Un second regroupement réduit à 9 le nombre de classes et à 36 887 le nombre d'objets, ce qui se rapproche d'une cartographie des «écosystèmes » de l'inventaire écologique. Sur le terrain, 315 relevés photographiques de la végétation ont été réalisés dans les vallées Akshayuk et Naqsaq. Pour chaque relevé, les pourcentages de couverture de 5 strates végétales ont été estimés, de manière à ranger les relevés dans 8 classes de végétation connues a priori. Dans l'image, ce sont 135 segments qui ont pu être retenus comme échantillons. De ce nombre, 71 et 64 échantillons furent retenus respectivement pour l'entraînement et la validation de la classification au plus proche voisin qui fut réalisée pour la végétation. L'exactitude générale de la classification de la végétation a été estimée à 54,7 %. Contrairement à la végétation, le couvert non-végétal est classifié suivant principalement un système de règles, lesquelles décrivent le comportement spectral de 34 types de surfaces selon une structure de classification hiérarchique.La classification des 20 écotypes non-végétaux a été validée par photo-interprétation à l'aide de 992 segments-non-végétale est évaluée à 83,2 %. Une fois synthétisée au niveau des écosystèmes, la classification atteint un taux de succès global de 92,7 %. Pour la classification de la végétation, l'analyse d'image par objets spatiaux livre une cartographie dont l'exactitude est équivalente à celle d'une classification basée sur le pixel réalisée par Parcs Canada pour la même image (54,7 % vs 53,4 %). Notre stratification comporte cependant un plus grand nombre de catégories non-végétales et leur classification atteint un niveau d'exactitude supérieur. L'analyse par objet spatiaux nous a permis d'aller au-delà de l'analyse pûrement spectrale pour incorporer des paramètres texturaux, géométriques et contextuels à la procédure de classification. Elle résulte en une représentation plus synthétique de l'information cartographique que la classification basée sur le pixel, mais les patrons spatiaux les plus fins des milieux les plus hétérogènes sont alors perdus.La structure de classification développée pour notre image peut être transposée avec succès vers une nouvelle image, mais ceci exige que soient apportés des ajustements aux règles de classification, voire l'ajout ou la suppréssion de certaines règles.La segmentation hiérarchique s'avère utile comme analogue au concept de l'inventaire écologique des parcs nationaux. Les informations véhiculées par chacun des niveaux de notre classification sont des intrants importants pour l'inventaire écologique du parc national du Canada Auyuittuq. Une typologie définitive reste à définir tant pour la classification de la végétation que pour le couvert non-végétal des parcs nationaux de l'Arctique. Des clés de classification seraient requises pour traduire ces typologies en paramètres reconnaissables sur le terrain. En mettant en commun les diverses données de terrain existantes pour le parc national du Canada Auyuittuq et en les structurant selon ces typologies, on obtiendrait une banque d'échantillons augmentée et plus cohérente. De telles données de références s'avéreraient une base solide pour la validation des classifications présentement disponibles ainsi que pour la mise en oeuvre de travaux futurs en matière d'inventaire écologique pour le parc national du Canada Auyuittuq.

Image compression based on Region of Interest (ROI) has been one of the hot topics of interest in image processing. There is not a single widely accepted method for detecting the ROI automatically form an image. To reduce the transmission bandwidth and storage space requirements of gray scale images, an algorithm is suggested for detecting the ROI automatically based on Tsallis entropy method. Tsallis entropy method is used to segment the image into two segments, the ROI and the background. These two segments can then be compressed at different rates, to avoid losing information in the ROI while achieving a good compression. Different approaches of compression based on wavelets and use of various compression methods are also discussed.

nÃo hà / The high inflationary indices that had lasted in Brazil in the 80âs, had been damaging, for several segments of the society, benefiting few economic sectors. Among those segments that had grown in this uncommon conjuncture, the financial institutions are distinguished, companies that focused their activities in terms of keeping public founds and application from federal governmentâs papers, avoiding to carry through operations of loans and financings, preventing to run in risk. The management used to maximize profits and to prevent risks took these institutions to carry through great investments in technological solutions and the main goal was to transform the processes of finance transferences into a very quick way and reduce the costs with salaries. As a consequence of this process the contact with clients became very impersonal. The view of knowing and evaluating the customers started right after the implantation of the Brazilian Real Plan was initiated after and beginning of the process of reduction of the inflation indices and taxes of interests. In this environment appeared the union corporations that work among doctors and belong to Unicred System, institutions that in last fifteen years comes conquering spaces in niches, that previously were exclusive for banks in commercial field and that encloses 129 cooperatives in Brazil. This paper presents an study about Unicred of Fortaleza, involving its controllers e cooperated clients, beyond the controllers of the Unicred Porto Alegre, Unicred of Southwestern Goias, Unicred of Southern Rio de Janeiro, Unicred Middle Plateaus and Unicred Saint Carlos, what are the only ones in Unicredâs System that work exclusively in the niche of doctors and their relatives. The work identify, in the perspective of the involved companies: the reasons from what doctors decided to belong to Unicred and maintain financial relationship with it, although all sort of financial institutions that can be found in the market; the view that they have of the services provided by their business; the importance of the main tools of marketing relationship used to conquer and maintain the associated partners and focus the risks of the competition of the banks over the niches in what the Unicred system what on. The study contributes in terms of making a review of the concept that doctors, in function of the image of a joined and cooperative class, have a strong cooperative feeling, being this the main booster of the growth of Unicred System. / Os elevados Ãndices inflacionÃrios que perduraram no Brasil na dÃcada de 1980, foram malÃficos para diversos segmentos da sociedade, beneficiando poucos setores econÃmicos. Entre os segmentos que cresceram nessa conjuntura incomum, destacam-se as instituiÃÃes financeiras, que focaram suas atividades na captaÃÃo de recursos junto ao pÃblico e aplicaÃÃo em tÃtulos do Governo Federal, evitando realizar operaÃÃes de emprÃstimos e financiamento para nÃo correrem riscos. As polÃticas de maximizar ganhos na captaÃÃo de recursos e evitar riscos nas operaÃÃes de crÃdito levaram essas instituiÃÃes a realizar elevados investimentos em soluÃÃes tecnolÃgicas, com o objetivo de tornar Ãgeis as transferÃncias de recursos, reduÃÃo dos custos operacionais e de mÃo-de-obra. Como conseqÃÃncia, o contato com a clientela tornou-se impessoal. A visÃo de obter resultados por meio do conhecimento e valorizaÃÃo do cliente iniciou-se apÃs a implantaÃÃo do Plano Real e da reduÃÃo dos Ãndices de inflaÃÃo e taxas de juros, que forÃaram os bancos a buscar novos ganhos na intermediaÃÃo financeira. Neste clima conturbado, surgiram as cooperativas de crÃdito, que atuam no segmento mÃdico associadas ao Sistema Unicred, instituiÃÃes que, nos Ãltimos quinze anos, conquistaram espaÃos em nichos anteriormente exclusivos dos bancos comerciais e que atualmente abrange 129 cooperativas em todo o Brasil. Esta dissertaÃÃo apresenta estudo realizado junto à Unicred Fortaleza, envolvendo seus cooperados e dirigentes, alÃm dos dirigentes da Unicred Porto Alegre, Sudoeste Goiano, Sul Fluminense, Planalto MÃdio e Unicred SÃo Carlos, que, a exemplo de Fortaleza, sÃo atualmente as cooperativas integrantes do Sistema Unicred a atuar, exclusivamente, com o nicho que abrange os profissionais mÃdicos e seus componentes familiares. O estudo identifica na perspectiva dos envolvidos: os motivos que levam os profissionais mÃdicos a se tornarem cooperados e a manterem relacionamento financeiro com a Unicred Fortaleza, apesar da diversidade de opÃÃes de instituiÃÃes financeiras existentes no mercado; a visÃo dos cooperados e dirigentes sobre os serviÃos ofertados por suas cooperativas; as principais ferramentas do marketing de relacionamento utilizadas na conquista e manutenÃÃo de cooperados; e destaca a ameaÃa da concorrÃncia dos bancos sobre o nicho de mercado atendido pelas cooperativas integrantes do Sistema Unicred. O estudo contribui para revisÃo do conceito de que os mÃdicos, em funÃÃo da imagem de classe unida e corporativista, teriam um forte sentimento cooperativista, sendo este o principal impulsionador do crescimento do Sistema Unicred.

O planejamento estratégico direcionado para o produto habitacional deve ser orientado para atender expectativas de um determinado estrato de mercado, apresentando características e atributos que satisfaçam à demanda desse público alvo nele inserido. Foram desenvolvidas diretrizes para estratificar um mercado habitacional e extrair informações sobre imagens de expectativas determinantes para a compra de um imóvel. O referencial teórico construído discutiu formas de estratificar um mercado por critérios geográficos, demográficos, socioeconômicos e psicográficos ou comportamentais. Identificaram-se expectativas habitacionais a partir do atendimento de necessidades durante o processo que conduziu a decisão de compra. A enquete realizada desenvolveu um processo de estratificação de mercado a partir da caracterização de um público alvo comprador de alto poder aquisitivo dentro de uma região da cidade de São Paulo. Este estrato, também foi produzido segundo os estágios do ciclo de vida familiar definidos, expresso por gerações e por um perfil psicográfico específico identificado também pela enquete. Ao final, as imagens de expectativas identificadas a partir do segmento de mercado produzido, emularam um produto hipotético apresentado que as atendesse de forma mais eficiente. / Strategic planning directed to the housing product should be directed to meet expectations of a certain stratum of the market, presenting features and attributes that meet the demand of target inserted therein. Guidelines have been developed to stratify a housing market and extract information about images determinants of expectations for the purchase of a property. The theoretical framework discussed ways to build a market stratification by geographic criteria, demographic, socioeconomic and psychographic or behavioral. Expectations were identified from the residential care needs during the process that led to the buying decision. The poll has developed a process of stratification of the market from the characterization of a buyer audience with high purchasing power within a region of São Paulo. This stratum was also produced according to the stages of family life cycle defined for generations and expressed by a specific psychographic profile also identified by the survey. In the end, the images from the expectations identified in the industry produced, emulated a product that presented hypothetical met more efficiently.

Neste trabalho é proposta uma metodologia para reconhecimento de anseios de segmento do mercado de real estate residencial. Será desenvolvido um protocolo contemplando um projeto de pesquisa para a formação de um banco de dados e o tratamento adequado das informações. O projeto de pesquisa parte do desenvolvimento de um questionário seguindo orientações científicas, fundamentado nas teorias de segmentação e nos fatores psicodinâmicos de influência do consumidor, resultando na estruturação de um banco de dados com as informações sobre o perfil dos entrevistados e seus anseios sobre imóveis residenciais. Aborda ainda o tratamento necessário das informações, que resultam no reconhecimento procurado para que sirva como ferramenta de ajuda na formatação de empreendimentos residenciais que utilizam segmentação como estratégia de inserção. Para validação do protocolo foi realizada uma pesquisa de caráter exploratório, aplicando o questionário resultante deste trabalho em 50 interessados na compra de imóveis residenciais. A amostra é formada de clientes que procuram imóveis nas regiões sul e sudeste da cidade em duas grandes imobiliárias da região. Os resultados obtidos na pesquisa formaram um banco de dados, deste foram escolhidos dois segmentos para análise de anseios: I) agrupados por ciclo de vida familiar: segmento de solteiros e II) segmento de clientes que procuram imóveis: casal com filhos. Os resultados foram comparados entre si e no total da amostra, possibilitando reconhecer os anseios de cada segmento e comprovando a importância da ferramenta proposta. / In this work the proposal is a methodology for recognition of yearnings of the residential market in real state. A protocol will be developed regarding to a research project for the formatting of a data base and the adjusted treatment of information. The research project goes from the development of a questionnaire following scientific directions, based on the recital theories and in the psychodynamics factors that influences the consumer. Resulting the formatting of a database with information about the interviewed profile and his yearnings for residential properties. It still approaches the necessary treatment of approaches that results in the looked recognition so that serves as an aid tool in the formatting of residential enterprises that use segmentation as insertion strategy. For validation of the protocol was realized an exploration character research applying the questionnaire of this work in 50 people interested in the purchase of residential enterprises. The sample is formed by costumers who look for properties in the regions south and southeastern of the city in two great real state ones of the regions. The gotten results in the research have formed a database; from these two segments had been chosen for earning analyses. i) grouped for cycle of familiar life: segment of bachelors, e ii) segment of: couples with children. The results where compared between it selves and with the total of the sample, making this way possible to recognize the yearnings of each segment and proving the importance of the proposed tool.

Multiple facets of perceived value perceptions drive loyalty intentions. However, this value-loyalty link is not uniform for all customers. In fact, the present study identifies three different segments that are internally consistent and stable across different service industries, using two data sets: the wireless telecommunication industry (sample size 1,122) and the financial services industry (sample size 982). Comparing the results of a single-class solution with finite mixture results confirms the existence of unobserved customer segments. The three segments found are characterized as "rationalists", "functionalists" and "value maximizers". These results point the way for value-based segmentation in loyalty initiatives and reflect the importance of a multidimensional conceptualization of perceived value, comprising cognitive and affective components. The present results substantiate the fact that assuming a homogeneous value-loyalty link provides a misleading view of the market. The paper derives implications for marketing research and practice in terms of segmentation, positioning, loyalty programs and strategic alliances. (authors' abstract)