Cette thèse propose d'explorer des méthodes de représentations parcimonieuses et d'apprentissage de dictionnaires pour compresser et classifier des images satellites. Les représentations parcimonieuses consistent à approximer un signal par une combinaison linéaire de quelques colonnes, dites atomes, d'un dictionnaire, et ainsi à le représenter par seulement quelques coefficients non nuls contenus dans un vecteur parcimonieux. Afin d'améliorer la qualité des représentations et d'en augmenter la parcimonie, il est intéressant d'apprendre le dictionnaire. La première partie de la thèse présente un état de l'art consacré aux représentations parcimonieuses et aux méthodes d'apprentissage de dictionnaires. Diverses applications de ces méthodes y sont détaillées. Des standards de compression d'images sont également présentés. La deuxième partie traite de l'apprentissage de dictionnaires structurés sur plusieurs niveaux, d'une structure en arbre à une structure adaptative, et de leur application au cas de la compression d'images satellites en les intégrant dans un schéma de codage adapté. Enfin, la troisième partie est consacrée à l'utilisation des dictionnaires structurés appris pour la classification d'images satellites. Une méthode pour estimer la Fonction de Transfert de Modulation (FTM) de l'instrument dont provient une image est étudiée. Puis un algorithme de classification supervisée, utilisant des dictionnaires structurés rendus discriminants entre les classes à l'apprentissage, est présenté dans le cadre de la reconnaissance de scènes au sein d'une image. / This thesis explores sparse representation and dictionary learning methods to compress and classify satellite images. Sparse representations consist in approximating a signal by a linear combination of a few columns, known as atoms, from a dictionary, and thus representing it by only a few non-zero coefficients contained in a sparse vector. In order to improve the quality of the representations and to increase their sparsity, it is interesting to learn the dictionary. The first part of the thesis presents a state of the art about sparse representations and dictionary learning methods. Several applications of these methods are explored. Some image compression standards are also presented. The second part deals with the learning of dictionaries structured in several levels, from a tree structure to an adaptive structure, and their application to the compression of satellite images, by integrating them in an adapted coding scheme. Finally, the third part is about the use of learned structured dictionaries for the classification of satellite images. A method to estimate the Modulation Transfer Function (MTF) of the instrument used to capture an image is studied. A supervised classification algorithm, using structured dictionaries made discriminant between classes during the learning, is then presented in the scope of scene recognition in a picture.

La presente investigación se desarrolla en torno a la problemática de pérdidas económicas y humanas recurrentes en áreas pobladas causadas por el flujo de escombros, conocido en Perú como “huayco” y sobre el cual hay poca información. Para evaluar el uso de las imágenes satelitales y SIG para calcular la producción de sedimentos (escombros) en el control de huaycos en tres quebradas de alto riesgo con pendientes que van del 29% al 35%. Para ello, era necesario obtener los volúmenes de erosión del suelo en la cuenca alta transportados por el flujo de escombros y los volúmenes retenidos por las barreras, especialmente para el ENSO 2017. Por lo que, se evaluaron las características geológicas, geomorfológicas e hidrológicas, así como la ocurrencia de eventos extremos; asimismo, se analizó el cambio geomorfológico con el uso del SIG e imágenes usando cartografía de los años 1960 y 1976 e imágenes satelitales del 2011 y 2013. Las investigaciones de campo posteriores al hecho del ENSO 2017 permitieron verificar que las mallas geodinámicas evitaron la pérdida de vidas humanas y daños materiales. No fue posible obtener resultados para todos los eventos destructivos porque no hay cartografía y las imágenes satelitales disponibles no tienen suficiente resolución temporal, espacial o presentan nubosidad mayor al 20%. Por lo que, se determinó que no es factible calcular la producción de sedimentos (escombros) en el control los huaycos en las quebradas en estudio mediante el uso de imágenes satelitales y el SIG. / This research is based on the recurring problem of economic and human losses occur in populated areas caused by the debris flow, known in Peru as “huayco” and for which there is little information. To evaluate the use of satellite images and Geographical Information System (GIS) to calculate sediment (debris) production in three high-risk creeks with slopes ranging from 29% to 35%. For that, it is necessary to obtain the volumes of both soil erosion in the upper basin transported by the debris flow and the retained solids by the barriers. Therefore, the geological, geomorphological, and hydrological characteristics were evaluated, as the occurrence of extreme events; Likewise, the geomorphological change was analyzed with the use of the GIS and images using cartography from the years 1960 and 1976 and satellite images from 2011 and 2013. It was no possible to obtain results for all the destructive events because there are no cartographies and the available satellite images do not have enough temporal or spatial resolution or present cloudiness greater than 20%. Therefore, it was determined that it is not feasible to calculate sediment (debris) production to control the huaycos in the streams under study using satellite images and the GIS. / Tesis

The aim of the project is to perform object based change detection of land cover in specific areas of Czech republic. Landsat 2000 and Spot 2006 satellite images are used as input data. The method used for evaluation of changes is Multivariate Alteration Detection unsupervised method which is based on statistical procedures and is available from e-Cognition software. The results of detection are compared with Corine Land Cover changes database to evaluate degree of parity on detected areas. Different mapping unit is used to be able to detect smaller changes than Corine database. First part of the work is review of literature sources aimed on processing of satellite images, description of the spectral behavior of landscape objects, origins of Corine Land Cover database and principle of change detection using MAD. Second part deals with data adjustment, change detection process and comparison of reached results with Corine. Keywords: object based change detection, satellite images, Corine Land Cover, mapping unit of changes, Multivariate Alteration Detection, e-Cognition

Les base de données (BD) d'Occupation du Sol (OCS) sont d'une grande utilité, dans divers domaines. Les utilisateurs recherchent des niveaux de détails tant géométriques que sémantiques très fins. Ainsi, une telle BD d'OCS à Grande Échelle (OCS-GE) est en cours de constitution à l'IGN. Cependant, pour répondre aux besoins des utilisateurs, cette BD doit être mise à jour le plus régulièrement possible, avec une notion de millésime. Ainsi, des méthodes automatiques de mise à jour doivent être mises en place, afin de traiter rapidement des zones étendues. Par ailleurs, les satellites d'observation de la terre ont fait leurs preuves dans l'aide à la constitution de BD d'OCS à des échelles comparables à celle de CLC. Avec l'arrivée de nouveaux capteurs THR, comme celle du satellite Pléiades, la question de la pertinence de ces images pour la mise à jour de BD d'OCS-GE se pose naturellement. Ainsi, l'objet de cette thèse est de développer une méthode automatique de mise à jour de BDs d'OCS-GE, à partir d'une image satellite THR monoscopique (afin de réduire les coûts d'acquisition), tout en garantissant la robustesse des changements détectés. Le cœur de la méthode est un algorithme d'apprentissage supervisés multi-niveaux appelé MLMOL, qui permet de prendre en compte au mieux les apparences, éventuellement multiples, de chaque thème de la BD. Cet algorithme, complètement indépendant du choix du classifieur et des attributs extraits de l'image, peut être appliqué sur des jeux de données très variés. De plus, la multiplication de classifications permet d'améliorer la robustesse de la méthode, en particulier sur des thèmes ayant des apparences multiples (e,g,. champs labourés ou non, bâtiments de type maison ou hangar industriel, ...). De plus, l'algorithme d'apprentissage est intégré dans une chaîne de traitements (LUPIN) capable, d'une part de s'adapter automatiquement aux différents thèmes de la BD pouvant exister et, d'autre part, d'être robuste à l'existence de thèmes in-homogènes. Par suite, la méthode est appliquée avec succès à une image Pléiades, sur une zone à proximité de Tarbes (65) couverte par la BD OCS-GE constituée par IGN. Les résultats obtenus montrent l'apport des images Pléiades tant en terme de résolution sub-métrique que de dynamique spectrale. D'autre part, la méthode proposée permet de fournir des indicateurs pertinents de changements sur la zone. Nous montrons par ailleurs que notre méthode peut fournir une aide précieuse à la constitution de BD d'OCS issues de la fusion de différentes BDs. En effet, notre méthode a la capacité de prise de décisions lorsque la fusion de BDs génère des zones de recouvrement, phénomène courant notamment lorsque les données proviennent de différentes sources, avec leur propre spécification. De plus, notre méthode permet également de compléter d'éventuels lacunes dans la zone de couverture de la BD générée, mais aussi d'étendre cette couverture sur l'emprise d'une image couvrant une étendue plus large. Enfin, la chaîne de traitements LUPIN est appliquée à différents jeux de données de télédétection afin de valider sa polyvalence et de juger de la pertinence de ces données. Les résultats montrent sa capacité d'adaptation aux données de différentes résolutions utilisées (Pléiades à 0,5m, SPOT 6 à 1,5m et RapidEye à 5m), ainsi que sa capacité à utiliser les points forts des différents capteurs, comme par exemple le canal red-edge de RapidEye pour la discrimination du thème forêts, le bon compromis de résolution que fournit SPOT 6 pour le thème zones bâties et l'apport de la THR de Pléiades pour discriminer des thèmes précis comme les routes ou les haies. / Land-Cover geospatial databases (LC-BDs) are mandatory inputs for various purposes such as for natural resources monitoring land planning, and public policies management. To improve this monitoring, users look for both better geometric, and better semantic levels of detail. To fulfill such requirements, a large-scale LC-DB is being established at the French National Mapping Agency (IGN). However, to meet the users needs, this DB must be updated as regularly as possible while keeping the initial accuracies. Consequently, automatic updating methods should be set up in order to allow such large-scale computation. Furthermore, Earth observation satellites have been successfully used to the constitution of LC-DB at various scales such as Corine Land Cover (CLC). Nowadays, very high resolution (VHR) sensors, such as Pléiades satellite, allow to product large-scale LC-DB. Consequently, the purpose of this thesis is to propose an automatic updating method of such large-scale LC-DB from VHR monoscopic satellite image (to limit acquisition costs) while ensuring the robustness of the detected changes. Our proposed method is based on a multilevel supervised learning algorithm MLMOL, which allows to best take into account the possibly multiple appearances of each DB classes. This algorithm can be applied to various images and DB data sets, independently of the classifier, and the attributes extracted from the input image. Moreover, the classifications stacking improves the robustness of the method, especially on classes having multiple appearances (e.g., plowed or not plowed fields, stand-alone houses or industrial warehouse buildings, ...). In addition, the learning algorithm is integrated into a processing chain (LUPIN) allowing, first to automatically fit to the different existing DB themes and, secondly, to be robust to in-homogeneous areas. As a result, the method is successfully applied to a Pleiades image on an area near Tarbes (southern France) covered by the IGN large-scale LC-DB. Results show the contribution of Pleiades images (in terms of sub-meter resolution and spectral dynamics). Indeed, thanks to the texture and shape attributes (morphological profiles, SFS, ...), VHR satellite images give good classification results, even on classes such as roads, and buildings that usually require specific methods. Moreover, the proposed method provides relevant change indicators in the area. In addition, our method provides a significant support for the creation of LC-DB obtain by merging several existing DBs. Indeed, our method allows to take a decision when the fusion of initials DBs generates overlapping areas, particularly when such DBs come from different sources with their own specification. In addition, our method allows to fill potential gaps in the coverage of such generating DB, but also to extend the data to the coverage of a larger image. Finally, the proposed workflow is applied to different remote sensing data sets in order to assess its versatility and the relevance of such data. Results show that our method is able to deal with such different spatial resolutions data sets (Pléiades at 0.5 m, SPOT 6 at 1.5 m and RapidEye at 5 m), and to take into account the strengths of each sensor, e.g., the RapidEye red-edge channel for discrimination theme forest, the good balance of the SPOT~6 resolution for built-up areas classes and the capability of VHR of Pléiades images to discriminate objects of small spatial extent such as roads or hedge.

Cette thèse considère le problème de sélection et d’ordonnancement des prises de vue d’un satellite agile d’observation de la Terre. La mission d’un satellite d’observation est d’obtenir des photographies de la surface de la Terre afin de satisfaire des requêtes d’utilisateurs. Les demandes, émanant de différents utilisateurs, doivent faire l’objet d’un traitement avant transmission d’un ordre vers le satellite, correspondant à une séquence d’acquisitions sélectionnées. Cette séquence doit optimiser deux objectifs sous contraintes d’exploitation. Le premier objectif est de maximiser le profit global des acquisitions sélectionnées. Le second est d’assurer l’équité du partage des ressources en minimisant la différence maximale de profit entre les utilisateurs. Deux métaheuristiques, composées d’un algorithme génétique à clé aléatoire biaisées (biased random key genetic algorithm - BRKGA) et d’une recherche locale multi-objectif basée sur des indicateurs (indicator based multi-objective local search - IBMOLS), sont proposées pour résoudre le problème. Pour BRKGA, trois méthodes de sélection, empruntées à NSGA-II, SMS-EMOA, et IBEA, sont proposées pour choisir un ensemble de chromosomes préférés comme ensemble élite. Trois stratégies de décodage, parmi lesquelles deux sont des décodages uniques et la dernière un décodage hybride, sont appliquées pour décoder les chromosomes afin d’obtenir des solutions. Pour IBMOLS, plusieurs méthodes pour générer la population initiale sont testées et une structure de voisinage est également proposée. Des expériences sont menées sur des cas réalistes, issus d’instances modifiées du challenge ROADEF 2003. On obtient ainsi les fronts de Pareto approximés de BRKGA et IBMOLS dont on calcule les hypervolumes. Les résultats de ces deux algorithmes sont comparés / This thesis considers the selection and scheduling problem of observations for agile Earth observing satellites. The mission of Earth observing satellites is to obtain photographs of the Earth surface to satisfy user requirements. Requests from several users have to be managed before transmitting an order, which is a sequence of selected acquisitions, to the satellite. The obtained sequence must optimize two objectives under operation constraints. The first objective is to maximize the total profit of the selected acquisitions. The second one is to ensure the fairness of resource sharing by minimizing the maximum profit difference between users. Two metaheuristic algorithms, consisting of a biased random key genetic algorithm (BRKGA) and an indicator-based multi-objective local search (IBMOLS), are proposed to solve the problem. For BRKGA, three selection methods, borrowed from NSGA-II, SMS-EMOA, and IBEA, are proposed to select a set of preferred chromosomes to be the elite set. Three decoding strategies, which are two single decoding and a hybrid decoding, are applied to decode chromosomes to become solutions. For IBMOLS, several methods for generating the initial population are tested and the neighborhood structure according to the problem is also proposed. Experiments are conducted on realistic instances based on ROADEF 2003 challenge instances. Hypervolumes of the approximate Pareto fronts are computed and the results from the two algorithms are compared

With the advancement of technology and the development of more sophisticated remote sensing sensor systems, the use of satellite imagery has opened up various fields of exploration and application. There has been an increased interest in analysis of multi-temporal satellite image in the past few years because of the wide variety of possible applications of in both short-term and long-term image analysis. The type of changes that might be of interest can range from short-term phenomena such as flood assessment and crop growth stage, to long-term phenomena such as urban fringe development. This thesis studies flood assessment and land cover mapping of satellite images, and proposes nature inspired algorithms that can be easily implemented in realistic scenarios. Disaster monitoring using space technology is one of the key areas of research with vast potential; particularly flood based disasters are more challenging. Every year floods occur in many regions of the world and cause great losses. In order to monitor and assess such situations, decision-makers need accurate near real-time knowledge of the field situation. How to provide actual information to decision-makers for effective flood monitoring and mitigation is an important task, from the point of view of public welfare. Over-estimation of the flooded area leads to over-compensation to people, while under-estimation results in production loss and negative impacts on the population. Hence it is essential to assess the flood damage accurately, both in qualitative and quantitative terms. In such situations, land cover maps play a very critical role. Updating land cover maps is a time consuming and costlier operation when it is performed using traditional or manual methods. Hence, there is a need to find solutions for such problem through automation. Design of automatic systems dedicated to satellite image processing which involves change detection to discriminate areas of land cover change between imaging dates. The system integrates the spectral and spatial information with the techniques of image registration and pattern classification using nature inspired techniques. In the literature, various works have been carried out for solving the problem of image registration and pattern classification using conventional methods. Many researchers have proved, for different situations, that nature inspired techniques are promising in comparison with that of conventional methods. The main advantage of nature inspired technique over any other conventional methods is its stochastic nature, which converges to optimal solution for any dynamic variation in a given satellite image. Results are given in such terms as to delineate change in multi-date imagery using change-versus-no-change information to guide multi-date data analysis. The main objective of this study is to analyze spatio-temporal satellite data to bring out significant changes in the land cover map through automated image processing methods. In this study, for satellite image analysis of flood assessment and land cover mapping, the study areas and images considered are: Multi-temporal MODerate-resolution Imaging Spectroradiometer (MODIS) image around Krishna river basin in Andhra Pradesh India; Linear Imaging Self Scanning Sensor III (LISS III)and Synthetic Aperture Radar(SAR)image around Kosi river basin in Bihar, India; Landsat7thematicmapperimage from the southern part of India; Quick-Bird image of the central Bangalore, India; Hyperion image around Meerut city, Uttar Pradesh, India; and Indian pines hyperspectral image. In order to develop a flood assessment framework for this study, a database was created from remotely sensed images (optical and/or Synthetic Aperture Radar data), covering a period of time. The nature inspired techniques are used to find solutions to problems of image registration and pattern classification of a multi-sensor and multi-temporal satellite image. Results obtained are used to localize and estimate accurately the flood extent and also to identify the type of the inundated area based on land cover mapping. The nature inspired techniques used for satellite image processing are Artificial Neural Network (ANN), Genetic Algorithm (GA),Particle Swarm Optimization (PSO), Firefly Algorithm(FA),Glowworm Swarm Optimization(GSO)and Artificial Immune System (AIS). From the obtained results, we evaluate the performance of the methods used for image registration and pattern classification to compare the accuracy of satellite image processing using nature inspired techniques. In summary, the main contributions of this thesis include (a) analysis of flood assessment and land cover mapping using satellite images and (b) efficient image registration and pattern classification using nature inspired algorithms, which are more popular than conventional optimization methods because of their simplicity, parallelism and convergence of the population towards the optimal solution in a given search space.

<p>У оквиру докторске дисертације извршен је преглед стања постојећих начина<br />идентификације геопросторних објеката на основу података насталих на принципима<br />даљинске детекције. Извршена је анализа постојећих проблема и корака које је<br />неопходно провести како би се добили што бољи резултати идентификације<br />геопросторних објеката. Анализирани су поступци мапирања, начини сегментације<br />слике, критеријуми за идентификацију, селекцију и класификацију геопросторних<br />објеката као и методе класификације. На основу анализе креиран је модел<br />идентификовања геопросторних објеката базираних на објектно оријентисаној анализи<br />слике. На основу предложеног модела извршена је верификација модела у поступку<br />идентификовања зграда, пољопривредних површина, шумских површина и водених<br />површина које представљају студије случаја.</p> / <p>U okviru doktorske disertacije izvršen je pregled stanja postojećih načina<br />identifikacije geoprostornih objekata na osnovu podataka nastalih na principima<br />daljinske detekcije. Izvršena je analiza postojećih problema i koraka koje je<br />neophodno provesti kako bi se dobili što bolji rezultati identifikacije<br />geoprostornih objekata. Analizirani su postupci mapiranja, načini segmentacije<br />slike, kriterijumi za identifikaciju, selekciju i klasifikaciju geoprostornih<br />objekata kao i metode klasifikacije. Na osnovu analize kreiran je model<br />identifikovanja geoprostornih objekata baziranih na objektno orijentisanoj analizi<br />slike. Na osnovu predloženog modela izvršena je verifikacija modela u postupku<br />identifikovanja zgrada, poljoprivrednih površina, šumskih površina i vodenih<br />površina koje predstavljaju studije slučaja.</p> / <p>This PhD thesis includes an overview of the existing methods of identifying geospatial<br />objects from a remote sensing data, basically satellite or airplane images. The analysis<br />of existing problems and necessary steps in identification of remotely sensed data is<br />obtained in way to get the best results of identification of geospatial objects. The<br />mapping procedures, methods of image segmentation, the criteria for identification,<br />selection and classification of geospatial objects and methods of classification are also<br />analyzed. The result of analysis is a model of identifying geospatial objects based on<br />object-oriented image analysis. Based on the proposed model, verification of the<br />model was carried out. Afterwards case study of the proposed model is carried out in<br />process of identifying buildings, farmland, forest and water areas.</p>

Mémoire numérisé par la Division de la gestion de documents et des archives de l'Université de Montréal.

As mudanças climáticas globais, regionais e locais representam, na atualidade, uma das maiores preocupações da humanidade. Essas mudanças podem ocorrer tanto a partir de causas naturais quanto a partir de causas antrópicas. As áreas das cidades se caracterizam por apresentarem temperaturas mais elevadas quando comparadas com as áreas rurais. Essa anomalia térmica causa a formação de ilhas de calor e esse fenômeno é reconhecidamente importante em estudos de clima urbano. O objetivo do presente trabalho foi, através de técnicas do sensoriamento remoto, identificar e analisar as ilhas de calor do Município de Piracicaba, SP verificando sua sazonalidade, intensidade e morfologia. Para tanto foi necessário realizar uma análise climática regional e verificar a possibilidade do uso do algoritmo de transformação termal do software IDRISI 3.2 nas imagens do satélite Landsat 7. Para validar o algoritmo foram aplicados dois métodos de transformação de temperatura aparente de superfície. Para a análise climática regional foram estudados os principais elementos climáticos do Município de Piracicaba, SP utilizando-se de dados da Estação Meteorológica da ESALQ/USP entre os anos de 1950 e 2005 e estes foram correlacionados com variáveis da urbanização. Concluiu-se, com os dados encontrados, que os elementos temperatura, precipitação, umidade relativa e evaporação tiveram tendência de aumento no período estudado e todos eles foram classificados como tendências climáticas. A temperatura apresentou tendência de aumento mais acentuada e se correlacionou positivamente com o aumento da urbanização. O algoritmo de transformação do software IDRISI 3.2 para o satélite Landsat 7 foi validado, sendo uma importante ferramenta para a utilização de imagens de melhor resolução. As ilhas de calor mais intensas do verão são representadas por locais com excesso de material de construção civil e pouca ou nenhuma área verde. A diferença entre a área urbana e a área rural da cidade ultrapassou 16°C no verão. O Parque da Rua do Porto é uma ilha de frescor e exerce um "efeito oásis" no centro e nos bairros vizinhos. O perfil das ilhas de calor do Município de Piracicaba não segue aquele delimitado por OKE (1974). As ilhas de calor variam sazonal e espacialmente e a intensidade destas, ao longo das estações do ano, está intimamente relacionada com a sazonalidade da cultura da cana-deaçúcar. As ilhas de calor da época da entressafra são, em média, 3.5°C mais intensas que as da época da safra. Por fim, pode-se afirmar que o uso e a cobertura do solo rural e urbano é um dos grandes agentes modificadores do clima local e regional. / Global, regional and local climate changes represent one of the greatest concerns of humanity. Climate changes can occur through natural or anthropogenic causes. Urban areas usually present higher temperatures than rural areas. This thermal effect is called "heat-island phenomenon" and has great importance on urban climate studies. In the present work, we identified and analyzed the heat-islands from Piracicaba, São Paulo using remote sensing techniques. The heat-islands were analyzed according to its seasonality, intensity and morphology using images from Landsat 7 satellite. We performed analysis on regional climate changes and investigated the use of the IDRISI thermal algorithm to convert Landsat 7 infrared thermal data on land surface temperature (LST). In order to transform Landsat 7 infrared thermal data we used two mathematical methods. Climate changes were analyzed by monitoring the climate elements for long periods of time, enabling the visualization of directional or periodical regional changes. The main climate elements were studied using data from ESALQ meteorological station for the last 55 years (1950-2005). Temperature, relative humidity, evaporation and precipitation variation were found to be correlated with urban growth parameters. The results indicated that temperature, precipitation, relative humidity and evaporation increased during the studied period and have been classified as "climate trends". The temperature presented the more accentuated trend of increase and was positively correlated with the growing urbanization. The software IDRISI 3.2 can be used with Landsat 7 high resolution images, being a useful and rapid tool to study urban heat islands. The most intense summer heatislands were represented by regions with higher amounts of constructed areas and almost any green area. In fact, during the summer the difference between the urban and rural areas was greater than 10°C. The Rua do Porto park was identified as a fresh-island and showed the "oasis effect" to the Center and neighbouring regions. Heat-islands varied according to the season and space and its intensity is intimately related to the sugar-cane seasonality. During the intercrop period the heat-islands were 3.5°C more intense than during the crop period. In conclusion land cover and land use affect local and regional climates.

Made available in DSpace on 2016-04-27T18:15:47Z (GMT). No. of bitstreams: 1 Israel da Silva Filho.pdf: 1290402 bytes, checksum: aeae05d6d347d992827f3dde615cf613 (MD5) Previous issue date: 2008-10-22 / Secretaria da Educação do Estado de São Paulo / This work is done to analyze and put in evidence some pedagogical procedures that can show us possibilities and limitations about using satellite images when teaching geography. It s so easy to find satellite images because they are all over the Internet sites but they haven t been used adequately by geography teachers because they don t know how to use it in order do make it easier for their students understand the space organization that they have to study. Besides, we think it s so important to discuss some questions related to pedagogical procedures, politics and economics that show us how the places can be so diverse when we are talking about new technology for teaching. In this case we have to consider the difficulty in accessing the internet, a poor structure in a public school, an over value attributed to new technology or even a refuse without any reasons, and the difficulty of working with this kind of resources. As satellites images are specific when compared with other traditional ways used to for teaching geography, we tried to show their specificity when we did and developed two activities students that are doing the seventh level of elementary school. At this time, that king of materials was used to show some aspects of the space that are impossible of been saw when we are using traditional ways of teaching. Then we tried to put in evidence that teachers have to be careful when using satellite images because they must use them connected with their pedagogical purposes which main subject is to prepare the students to read, to interpret and to transform their reality / Este trabalho analisa encaminhamentos didáticos que revelam possibilidades e limitações do uso das imagens de satélite no ensino de Geografia. O acesso relativamente fácil às imagens de satélite disponíveis em diferentes sites na Internet e a utilização desses recursos nas aulas de Geografia não tem recebido por parte do professor de Geografia um tratamento didático coerente que ajude os estudantes compreenderem a dimensão geográfica dos fenômenos observados. Aliado a este fato, há questões de ordem didática, política e econômica que revelam um cenário bastante diversificado quanto ao uso das Novas Tecnologias no ensino e merecem ser destacados: ausência ou baixos índices de acesso devido à estrutura precária dos estabelecimentos; valorização exacerbada das novas tecnologias como resolução de problemas na escola ou, em contra partida, repúdio gratuito e dificuldade de trabalhar com essas ferramentas. Como as imagens de satélite têm especificidades em relação a outros recursos didáticos tradicionalmente utilizados no ensino de Geografia, elas são ressaltadas no desenvolvimento de duas atividades elaboradas para as 7ªs séries do Ensino Fundamental para mostrar aspectos do espaço que são pouco visíveis nos mapas disponíveis na escola. No decorrer do trabalho são enfatizados alguns cuidados que os professores precisam tomar para que o uso das imagens de satélite seja feito em sintonia com a organização pedagógica do professor que cotidianamente prepara os alunos para a leitura, interpretação e transformação da realidade em que vivem