<p>Basic radar systems use electromagnetic wave reflections from targets to determine the motion characteristics of these targets. Synthetic Aperture Radar (SAR) systems use the reflections to produce target images as well. SAR is an imaging radar system that produces high resolution images of a scene or target by using radar motion to synthesize the antenna aperture. A SAR model to handle extended targets and point targets in faster time is presented, as are some simulated results. This thesis explains synthetic aperture concepts, the model used and a simulation of a SAR system. It runs through modelling point targets as well as extended targets by using the resolution cells of the radar, creating the raw signal data from the target information and then the signal processing that converts the raw data to a SAR image. The simulation was done for better understanding of synthetic aperture parameters and it was done in C++ programming language for improved processing speed. In comparison to previous simulations obtained from literature review, there is an increase in speed of more than 2.5 times as the number of targets increases, producing higher resolution images in less time. A model to handle extended targets was presented while also showing the imperfections due to the model assumptions. These assumptions are then explained as the best option in the absence of extra geographic information on the target scene.</p> / Master of Applied Science (MASc)

Throughout the years of space exploration, the usage of Earth observation satellites has increased tremendously. The usage today extends beyond optical sensors, encompassing radars, infrared, and laser sensors. For this thesis, the usage of optical, synthetic aperture radar, and LiDAR sensors were looked at to see if the fusion of these different sensors could enhance the overall image quality. A crucial aspect of satellite image fusion, regardless of sensor type, is preprocessing to ensure the individual images can be seamlessly merged. Ultimately these preprocessing steps are individual to both the sensors and even different satellites. The topic of remote sensing and satellite image fusion is extensive and complex. Therefore, this thesis aims to explore various fusion techniques, data sources, and algorithms to contribute to a deeper understanding of the advantages but mostly challenges associated with multisensor satellite image fusion. A web scraper was developed to collect data from the European Space Agency’s Third Party Mission website, a central repository for satellite missions and Earth images. The scraper made it possible to select different satellites and find the image areas which they had in common. A way to process this data is then presented on how to process the images and finally fuse them. The three fusion algorithms that were used were a simple weighted average, intensity hue saturation, and the pansharpening method. The pansharpening increased both the spatial and spectral resolution whereas the fusion of the optical and synthetic aperture radar gave some mixed results. There are a lot of things that could be explored in the future, such as utilizing more complex fusion algorithms or using additional satellite sensors. However, the web scraper and the processing flowchart stand as notable achievements of this thesis, simplifying the entire process of multisensor satellite image fusion. / Genom åren har användningen av jordobservationsatelliter ökat avsevärt inom rymdforskning. Användningen sträcker sig idag bortom optiska sensorer och inkluderar även radar-, infraröd- och lasersensorer. I detta examensarbete undersöks användningen av optiska, syntetisk aperturradar- och LiDARsensorer för att se om fusionen av dessa olika sensorer kan förbättra helhetsbilden av ett område. En avgörande aspekt av satellitbildsfusion, oavsett sensortyp, är förbehandling för att säkerställa att de individuella bilderna kan smidigt integreras. Slutligen är dessa förbehandlingsteg specifika för både sensorerna och olika satelliter. Ämnet fjärranalys och fusion av satellitbilder är omfattande och komplext. Därför syftar detta examensarbete till att utforska olika fusionsmetoder, datakällor och algoritmer för att bidra till en djupare förståelse för fördelarna och utmaningarna med multisensor fusion av satellitbilder. Ett av huvudproblemen under examensarbetet var datainsamling och databehandling. För att överkomma detta utvecklades en webbskrapare för att samla in data från European Space Agencys Third Party Mission hemsida, en central databas för satellituppdrag och bilder av jorden. Skrapan möjliggjorde valet av olika satelliter och identifieringen av gemensamma bildområden. En metod för databehandling presenteras sedan för att bearbeta bilderna och slutligen förena dem. De tre fusionsalgoritmerna som användes var en enkel viktad medelvärdesmetod, intensitetssättning och pansharpening. Pansharpening ökade både den spatiala och spektrala upplösningen, medan fusionen av optiska och syntetisk aperturradar gav blandade resultat. Det finns många områden som kan utforskas i framtiden, såsom användning av mer komplexa fusionsalgoritmer eller ytterligare satellitsensorer. Sammanfattande kan webbskrapan och behandlingsflödet ses som betydande framsteg i detta examensarbete och förhoppningsvis förenkla hela processen med multisensor fusion av satellitbilder.

近年全球暖化現象日益嚴重，格陵蘭等極區融冰所造成之海平面上升將對全球人類帶來嚴重威脅。因冰層質量之改變與冰河移動速度高度相關，故可藉由監測格陵蘭冰層（Greenland Ice Sheet，GrIS）上冰河之移動推估全球暖化對其造成之影響。衛星影像因具有連續且快速獲得大範圍地表資訊之能力，且可結合各影像處理技術獲得地表變形量，故已廣泛應用於廣域冰河之監測。然不同影像與技術均有其優勢與限制，故本研究將使用合成孔徑雷達（Synthetic Aperture Radar，SAR）與光學影像，並結合合成孔徑雷達差分干涉（Differential Interferometric SAR，D-InSAR）、多重合成孔徑雷達干涉（Multi-aperture Interferometric SAR，MAI）與偏移偵測法（Pixel-offset，PO）技術獲得冰河表面於不同方向之位移向量，再整合各向量透過三維變動量解構法（3D decomposition）求解表面於三維方向之變形量。據此執行數值冰層動力模型（Numerical Ice Sheet Model，ISM），並結合模擬之冰底基岩渠道網絡、數化之冰面冰隙與冰面湖及氣象觀測資料後，參佐冰河變動理論，進一步了解格陵蘭Russell冰河之變動行為與機制。 / Global warming has been a worldwide issue and significantly increasing icecap melting rate over polar area. Consequently the sea level rises continuously and poses a fundamental threat to whole human beings. Since the mass loss of Greenland ice sheet （GrIS） is highly correlated to the velocity of glacier movement, this study aims to monitor the impact of global warming by tracking glacier terminus displacement over GrIS using remote sensing techniques. As there are multiple spaceborne images of various characteristics and also multiple techniques with different functions, we proposed a monitoring strategy using Synthetic Aperture Radar （SAR） and optical images, with Differential Interferometric SAR （D-InSAR）, Multi-aperture Interferometric SAR （MAI） and Pixel-offset （PO） techniques to estimate glacier movement vectors. The vectors were then merged using 3D decomposition method to derive 3D deformation. Based on the resultant 3D deformation, the Numerical Ice Sheet Model （ISM） is conducted and then integrates with modeled subglacial drainage channel network and glaciological theories, the melting dynamics and mechanism of Russell glacier can be further understood.

Les radars à synthèse d’ouverture (RSO) polarimétriques sont devenus incontournables dans le domaine de la télédétection, grâce à leur zone de couverture étendue, ainsi que leur capacité à acquérir des données dans n’importe quelles conditions atmosphériques de jour comme de nuit. Au cours des trois dernières décennies, plusieurs RSO polarimétriques ont été utilisés portant une variété de modes d’imagerie, tels que la polarisation unique, la polarisation double et également des modes dits pleinement polarimétriques. Grâce aux recherches récentes, d’autres modes alternatifs, tels que la polarisation hybride et compacte, ont été proposés pour les futures missions RSOs. Toutefois, un débat anime la communauté de la télédétection quant à l’utilité des modes alternatifs et quant au compromis entre la polarimétrie double et la polarimétrie totale. Cette thèse contribue à ce débat en analysant et comparant ces différents modes d’imagerie RSO dans une variété d’applications, avec un accent particulier sur la surveillance maritime (la détection des navires et de marées noires). Pour nos comparaisons, nous considérons un paramètre fondamental, appelé le degré de polarisation (DoP). Ce paramètre scalaire a été reconnu comme l’un des paramètres les plus pertinents pour caractériser les ondes électromagnétiques partiellement polarisées. A l’aide d’une analyse statistique détaillée sur les images polarimétriques RSO, nous proposons des estimateurs efficaces du DoP pour les systèmes d’imagerie cohérente et incohérente. Ainsi, nous étendons la notion de DoP aux différents modes d’imagerie polarimétrique hybride et compacte. Cette étude comparative réalisée dans différents contextes d’application dégage des propriétés permettant de guider le choix parmi les différents modes polarimétriques. Les expériences sont effectuées sur les données polarimétriques provenant du satellite Canadian RADARSAT-2 et le RSO aéroporté Américain AirSAR, couvrant divers types de terrains tels que l’urbain, la végétation et l’océan. Par ailleurs nous réalisons une étude détaillée sur les potentiels du DoP pour la détection et la reconnaissance des marées noires basée sur les acquisitions récentes d’UAVSAR, couvrant la catastrophe de Deepwater Horizon dans le golfe du Mexique. / Polarimetric Synthetic Aperture Radar (SAR) systems have become highly fruitful thanks to their wide area coverage and day and night all-weather capabilities. Several polarimetric SARs have been flown over the last few decades with a variety of polarimetric SAR imaging modes; traditional ones are linear singleand dual-pol modes. More sophisticated ones are full-pol modes. Other alternative modes, such as hybrid and compact dual-pol, have also been recently proposed for future SAR missions. The discussion is vivid across the remote sensing society about both the utility of such alternative modes, and also the trade-off between dual and full polarimetry. This thesis contributes to that discussion by analyzing and comparing different polarimetric SAR modes in a variety of geoscience applications, with a particular focus on maritime monitoring and surveillance. For our comparisons, we make use of a fundamental, physically related discriminator called the Degree of Polarization (DoP). This scalar parameter has been recognized as one of the most important parameters characterizing a partially polarized electromagnetic wave. Based on a detailed statistical analysis of polarimetric SAR images, we propose efficient estimators of the DoP for both coherent and in-coherent SAR systems. We extend the DoP concept to different hybrid and compact SAR modes and compare the achieved performance with different full-pol methods. We perform a detailed study of vessel detection and oil-spill recognition, based on linear and hybrid/compact dual-pol DoP, using recent data from the Deepwater Horizon oil-spill, acquired by the National Aeronautics and Space Administration (NASA)/Jet Propulsion Laboratory (JPL) Uninhabited Aerial Vehicle Synthetic Aperture Radar (UAVSAR). Extensive experiments are also performed over various terrain types, such as urban, vegetation, and ocean, using the data acquired by the Canadian RADARSAT-2 and the NASA/JPL Airborne SAR (AirSAR) system.

Conselho Nacional de Desenvolvimento CientÃfico e TecnolÃgico / A cobertura regular de quase todo o planeta por sistemas de radar de abertura sintÃtica (synthetic aperture radar - SAR) orbitais e o uso de sistemas aerotransportados tÃm propiciado novos meios para obter informaÃÃes atravÃs do sensoriamento remoto de vÃrias regiÃes de nosso planeta, muitas delas inacessÃveis. Este trabalho trata do processamento de imagens digitais geradas por radar de abertura sintÃtica, especificamente da segmentaÃÃo, que consiste do isolamento ou particionamento dos objetos relevantes presentes em uma cena. A segmentaÃÃo de imagens digitais visa melhorar a interpretaÃÃo das mesmas em procedimentos subseqÃentes. As imagens SAR sÃo corrompidas por ruÃdo coerente, conhecido por speckle, que mascara pequenos detalhes e zonas de transiÃÃo entre os objetos. Tal ruÃdo à inerente ao processo de formaÃÃo dessas imagens e dificulta tarefas como a segmentaÃÃo automÃtica dos objetos existentes e a identificaÃÃo de seus contornos. Uma possibilidade para efetivar a segmentaÃÃo de imagens SAR consiste na filtragem preliminar do ruÃdo speckle, como etapa de tratamento dos dados. A outra possibilidade, aplicada neste trabalho, consiste em segmentar diretamente a imagem ruidosa, usando seus pixels originais como fonte de informaÃÃo. Para isso, à desenvolvida uma metodologia de segmentaÃÃo baseada em crescimento e fusÃo estatÃstica de regiÃes, que requer alguns parÃmetros para controlar o processo. As vantagens da utilizaÃÃo dos dados originais para realizar a segmentaÃÃo de imagens de radar sÃo a eliminaÃÃo de etapas de prÃ-processamento e o favorecimento da detecÃÃo das estruturas presentes nas mesmas. à realizada uma avaliaÃÃo qualitativa e quantitativa das imagens segmentadas, sob diferentes situaÃÃes, aplicando a tÃcnica proposta em imagens de teste contaminadas artificialmente com ruÃdo multiplicativo. Este segmentador à aplicado tambÃm no processamento de imagens SAR reais e os resultados sÃo promissores. / The regular coverage of the planet surface by spaceborne synthetic aperture radar (SAR)and also airborne systems have provided alternative means to gather remote sensing information of various regions of the planet, even of inaccessible areas. This work deals with the digital processing of synthetic aperture radar imagery, where segmentation is the main subject. It consists of isolating or partitioning relevant objects in a scene, aiming at improving image interpretation and understanding in subsequent tasks. SAR images are contaminated by coherent noise, known as speckle, which masks small details and transition zones among the objects. Such a noise is inherent in radar image generation process, making difficult tasks like automatic segmentation of the objects, as well as their contour identification. To segment radar images, one possible way is to apply speckle filtering before segmentation. Another one, applied in this work, is to perform noisy image segmentation using the original SAR pixels as input data, without any preprocessing,such as filtering. To provide segmentation, an algorithm based on region growing and statistical region merging has been developed, which requires some parameters to control the process. This task presents some advantages, as long as it eliminates preprocessing steps and favors the detection of the image structures, since original pixel information is exploited. A qualitative and quantitative performance evaluation of the segmented images is also executed, under different situations, by applying the proposed technique to simulated images corrupted with multiplicative noise. This segmentation method is also applied to real SAR images and the produced results are promising.

This study analyzes the available radar evidence in order to describe the morphology of channel networks around the north polar seas of Titan. Critical flow depths necessary to entrain water-ice grains, and denudation rates for a north polar channel network are discussed. The results indicate that channel networks on Titan have similar morphologies to channel networks cut by water on Earth. We also find that water-ice sediment should be readily entrained in the headwaters and downstream sections of the analyzed Titanian basin, given sufficient flow depths of liquid hydrocarbons. Also, the importance of slope and the elevated topography of the highlands surrounding the polar lakes are considered, as well as potential formation theories for the elevated highlands and low-lying maria that dominate the north polar region.

Submitted by Geyciane Santos (geyciane_thamires@hotmail.com) on 2015-07-27T15:15:30Z No. of bitstreams: 1 Dissertação - Jorge Alberto Lopes da Costa.pdf: 12254367 bytes, checksum: e92aa3fbce27b2b569e2f3aae45e851d (MD5) / Approved for entry into archive by Divisão de Documentação/BC Biblioteca Central (ddbc@ufam.edu.br) on 2015-07-28T15:14:18Z (GMT) No. of bitstreams: 1 Dissertação - Jorge Alberto Lopes da Costa.pdf: 12254367 bytes, checksum: e92aa3fbce27b2b569e2f3aae45e851d (MD5) / Approved for entry into archive by Divisão de Documentação/BC Biblioteca Central (ddbc@ufam.edu.br) on 2015-07-28T15:19:49Z (GMT) No. of bitstreams: 1 Dissertação - Jorge Alberto Lopes da Costa.pdf: 12254367 bytes, checksum: e92aa3fbce27b2b569e2f3aae45e851d (MD5) / Made available in DSpace on 2015-07-28T15:19:50Z (GMT). No. of bitstreams: 1 Dissertação - Jorge Alberto Lopes da Costa.pdf: 12254367 bytes, checksum: e92aa3fbce27b2b569e2f3aae45e851d (MD5) Previous issue date: 2011-07-07 / Não informada / In recent decades the areas of rainforest in the Amazon region has been heavily impacted by a rapid process of conversion of vegetation cover in other types of use due to human action. In the context of global change, the use of mapping and monitoring land cover and provide information for the analysis and evaluation of environmental impacts due to accelerated changes in the landscape. Therefore, this study evaluated the potential of data from synthetic aperture radar for discriminating use and land cover in the region of Manaus, Amazonas state. We used a multipolarized image from sensor airborne SAR-R99B (L band), with 3 m spatial resolution. Were evaluated the MAXVER-ICM and SVM (Support Vector Machine) classifiers, where in all cases we used the images individually multipolarized amplitude (HH, HV and VV), in pairs (HH and HV), (HV and VV) and (HH and VV) and together (HH, HV and VV). The results were compared using as parameter the Kappa coefficient. The SVM classifier had higher accuracy compared to MAXVER-ICM classifier. The best classifications were obtained for the dual polarization (HH and VV) with MARVER-ICM classifier and (HH, HV and VV) with the SVM classifier both using the images with the filter. The accuracy was highest with SVM for classification and filter images (kappa = 0.7736). Were analyzed the influence of using GAMMA filter performance on the classifiers where it showed that filtered images have provided an increase in the results, on average, about 8%. Thus there was the analysis of the classification results, which found that the best result was provided by the dataset multipolarized (HH, HV and VV) classified by the SVM method. Thus, we concluded that the use of radar imagery in mapping thematic classes use and land cover in tropical regions, can be considered as a viable proposal. / Nas últimas décadas as áreas de floresta tropical na região Amazônica têm sido fortemente impactada por um rápido processo de conversão da cobertura vegetal em outros tipos de uso devido à ação antrópica. No contexto das mudanças globais, os mapeamentos e monitoramentos de uso e cobertura da terra fornecem subsídios para as análises e avaliações dos impactos ambientas em virtude de acelerados processos de mudança na paisagem. Neste contexto, este estudo avaliou o potencial dos dados de radar de abertura sintética para discriminação de uso e cobertura da terra na região de Manaus, estado do Amazonas. Foi utilizada uma imagem multipolarizada do sensor aerotransportado SAR-R99B (banda L), com 3 metros de resolução espacial. Realizaram-se classificações na imagem radar sem filtro e com filtro Gamma 3x3. Avaliou-se o classificador pontual MAXVER-ICM e o SVM (Support Vector Machine), onde em todos os casos utilizou-se das imagens multipolarizadas em amplitude individualmente (HH, HV e VV), aos pares (HH e HV), (HV e VV) e (HH e VV) e em conjunto (HH, HV e VV). Os resultados obtidos foram comparados utilizando-se como parâmetro o coeficiente de concordância Kappa. O classificador SVM apresentou acurácia superior em relação ao classificador MAXVER-ICM. As melhores classificações foram obtidas para a polarização dual HH e VV com o classificador MAXVER-ICM e (HH, HV e VV) com o classificador SVM ambos utilizando as imagens com filtro. A acurácia mais elevada foi para a classificação com SVM e imagens com filtro (kappa = 0,7736). Analisou-se a influência do uso de filtro GAMMA no desempenho dos classificadores onde se contatou que as imagens filtradas proporcionaram um incremento nos resultados, em média, na ordem de 8%. Deste modo realizou-se a análise dos resultados das classificações, onde se constatou que o melhor resultado foi proporcionado pelo conjunto de dados multipolarizados (HH, HV e VV)classificados através do método SVM. Assim, concluiu-se que o uso de imagens de radar no mapeamento de classes temáticas de uso e cobertura da terra, em regiões tropicais, pode ser considerado como uma proposta viável.

L'urbanisation et la gestion de l'environnement urbain et sa périphérie deviennent l'un des problèmes les plus cruciaux dans les pays développés et en développement. Dans ces circonstances, les données de télédétection sont une source importante d'information qui reflète les interactions entre les êtres humains et leur environnement. Compte tenu de leur indépendance totale des contraintes logistiques sur le terrain, l'éclairage (lumière du jour) et météorologiques (nuages) conditions, Synthetic Aperture Radar (SAR) les systèmes de satellites peuvent fournir des contributions importantes dans des environnements complexes de reconstruction 3-D. La nouvelle génération de haute résolution SAR capteurs comme COSMOSkyMed, TerraSAR-X RADARSAT-2 a permis d'acquérir des images SAR à haute résolution. Ici, l'attention est mis sur la technique pour l'imagerie 3-D nominée tomographie SAR: à partir d'une pile d'images ont été recueillies en utilisant les données multibaseline effectuées dans la configuration interférométrique, une telle technique permet d'extraire les informations de hauteur formant une ouverture synthétique dans la direction d'élévation afin d'obtenir une résolution sensiblement améliorée. Cette thèse de doctorat se concentre sur les potentialités élevées de techniques tomographiques en 3-D surveillance des changements et la caractérisation des zones complexes et denses bâties en utilisant des estimateurs mono-dimensionnelle de base comme Beamforming, Capon et MUSIC combinée au satellite très haute résolution des images SAR. 2-D et de l'analyse 3-D ont été présentés sur la zone urbaine de Paris en utilisant les données TerraSAR-X à haute résolution et de polarisation unique. Être porté principalement sur les techniques tomographiques 3-D, dans les méthodes de travail 4-D présentés, tels que le Compressive Sensing, ne sont pas pris en compte. Dans un premier temps, l'analyse de la qualité interférométrique de l'ensemble de données transformées a montré de bonnes valeurs de cohérence moyenne et ont permis de détecter des images considérées comme des valeurs aberrantes. L'extraction des tomographies 2-D sur l'azimut différent de profil a montré la capacité de distinguer plus d'un diffuseur à l'intérieur de la même cellule de résolution et de reconstituer les profilés de construction verticaux. Successivement, une caractérisation 3-D globale en terme de bâtiments hauteurs et réflectivité verticale a été réalisée dans le but de développer un outil de suivi des changements des structures simples. En outre, la possibilité de corriger les distorsions géométriques en raison de l'escale (qui affecte fortement ce genre de scénarios) et de déterminer les informations sur le nombre de diffuseurs (jusqu'à trois) et la réflectivité correspondant à l'intérieur d'une cellule de résolution ont été évalués. / The urbanization and the management of urban environment and its periphery become one of the most crucial issues in developed and developing countries. In these circumstances, remote sensing data are an important source of information that reflects interactions between human beings and their environment. Given their complete independence from logistic constraints on the ground, illumination (daylight), and weather (clouds) conditions, Synthetic Aperture Radar (SAR) satellite systems may provide important contributions in complex environments 3-D reconstruction. The new generation of high resolution SAR sensors as COSMO-SkyMed, TerraSAR-X and RADARSAT-2 allowed to acquire high resolution SAR imagery. Here the attention is put on the 3-D imaging technique called SAR Tomography: starting from a stack of images collected using multibaseline data performed in interferometric configuration, such a technique allows to retrieve height information forming a synthetic aperture in the elevation direction in order to achieve a substantially improved resolution. The present PhD thesis is focused on the high potentialities of tomographic techniques in 3-D change monitoring and characterization for complex and dense built-up areas using basic mono-dimensional estimators as Beamforming, Capon and MUSIC combined to very high satellite SAR resolution imagery. 2-D and 3-D analysis have been presented over the urban area of Paris using TerraSAR-X data at high resolution and single polarisation. Being mainly focused on the 3-D tomographic techniques, in the presented work 4-D methods, such as compressive sensing (CS), have not been taken into account. At first, the analysis of the interferometric quality of the processed data set has been performed and results showed good mean coherence values within the entire stack. The extraction of 2-D tomograms over different azimuth-profile has showed the capabilities to distinguish more than one scatterer within the same resolution cell and to reconstruct the vertical building profiles. Successively, a global 3-D characterization both in term of buildings heights and vertical reflectivity has been performed in order to develop a monitoring tool for the changes of single structures. Moreover, the possibility to correct the geometric distortions due to the layover (that strongly affects such kind of scenarios) and to determine the information about the number of scatterers (up to three) and the corresponding reflectivity within one resolution cell have been evaluated. Moreover an innovative time stability analysis of the observed scene have been carried out in order to detect the stable and unstable scatterers. Globally, the investigations showed noisier and sparser point clouds for the Capon method, whereas better capabilities for the Beamforming and MUSIC ones. Indeed, it was possible to detect different scatterers located within the same resolution cell and to resolve pixels affected by the layover. This has lead to perform a good reconstruction of building shape and location and a good estimation of their elevation. The 3-D time stability analysis demonstrated the possibility to monitor the 3-D change depending on the time. Eventually, it is possible to assert that processing high resolution SAR data allows to achieve a strong improvement in 3-D imaging capabilities. It has been demonstrated the potentialities of TomoSAR technique in distortions correction and in 3-D change monitoring using basic mono-dimensional estimators.

La caractérisation de la neige est un enjeu important pour la gestion des ressources en eau et pour la prévision des risques d'avalanche. L'avènement des nouveaux satellites Radar de Synthèse d'Ouverture (RSO) bande X à haute résolution permet d'acquérir des données de résolution métrique avec une répétitivité journalière. Dans ce travail, un modèle de rétrodiffusion des ondes électromagnétiques de la neige sèche est adapté à la bande X et aux fréquences plus élevées. L'algorithme d'assimilation de données 3D-VAR est ensuite implémenté pour contraindre le modèle d'évolution de la neige SURFEX/Crocus à l'aide des observations satellitaires. Enfin, l'ensemble de ces traitements sont évalué à partir de données du satellite TerraSAR-X acquises sur le glacier d'Argentière dans la vallée de Chamonix. Cette première comparaison montre le fort potentiel de l'assimilation des données RSO bande X pour la caractérisation du manteau neigeux. / Characterization of snowpack structure is an important issue for the management of water resources and the prediction of avalanche risks. New Synthetic Aperture Radar (SAR) satellites in X-band at high-resolution allow us to acquire image data with metric resolution and daily observations. In this work, an electromagnetic backscattering model applicable for dry snow is adapted for X-band and higher frequencies. The 3D-VAR data assimilation algorithm is then implemented to constrain the evolution of the snow metamorphisme model SURFEX/Crocus using satellite observations. Finally, the algorithm is evaluated using image data acquired from TerraSAR-X satellite on the Argentiere glacier in the Chamonix Valley of the French Alps. This first comparison shows the high potential of the data assimilation assimilation method using X-band SAR data for characterization of the snowpack.

En raison de la nature cohérente du signal RADAR à synthèse d’ouverture (RSO), les images RSO polarimétriques (RSOPOL) sont affectées par le bruit de chatoiement. L’effet du chatoiement peut être sévère au point de rendre inutilisable la donnée RSOPOL. Ceci est particulièrement vrai pour les données à une vue qui souffrent d’un chatoiement très intense.Un filtrage du bruit est nécessaire pour améliorer l’estimation des paramètres polarimétriques pouvant être calculés à partir de ce type de données. Cette opération constitue une étape importante dans le traitement et l’analyse des images RSOPOL. Récemment une nouvelle approche est apparue en traitement de données visant la solution d’une multitude de problèmes dont le filtrage, la restauration d’images, la reconnaissance de la parole, la classification ou la segmentation d’images. Cette approche est l’apprentissage profond et les réseaux de neurones à convolution (RNC). Des travaux récents montrent que les RNC sont une alternative prometteuse pour le filtrages des images RSO. En effet par leur capacité d’apprendre un modèle optimal de filtrage, ils tendent à surpasser les approches classiques du filtrage sur les images RSO. L’objectif de cette présente étude est d’analyser et d’évaluer l’efficacité du filtrage par RNC sur des données RSOPOL simulées et sur des images satellitaires RSOPOL RADARSAT-2, ALOS/PalSAR et GaoFen-3 acquises sur la région urbaine de San Francisco (Californie). Des modèles inspirés de l’architecture d’un RNC utilisé notamment en Super-résolution ont été adaptés pour le filtrage de la matrice de cohérence polarimétrique. L’effet de différents paramètres structuraux de l’architecture des RNC sur le filtrage ont été analysés, parmi ceux-ci on retrouve entre autres la profondeur du réseau (le nombre de couches empilées), la largeur du réseau (le nombre de filtres par couches convolutives) et la taille des filtres de la première couche convolutive. L’apprentissage des modèles a été effectué par la rétropropagation du gradient de l’erreur en utilisant 3 ensembles de données qui simulent la polarimétrie une vue des diffuseurs selon les classes de Cloude-Pottier. Le premier ensemble ne comporte que des zones homogènes.Les deux derniers ensembles sont composés de simulations en patchwork dont l’intensité locale est simulée par des images de texture et de cibles ponctuelles ajoutées au patchwork dans le cas du dernier ensemble. Les performances des différents filtres par RNC ont été mesurées par des indicateurs comprenant l’erreur relative sur l’estimation de signatures polarimétriques et des paramètres de décomposition ainsi que des mesures de distorsion sur la récupération des détails importants et sur la conservation des cibles ponctuelles. Les résultats montrent que le filtrage par RNC des données polarimétriques est soit équivalent ou nettement supérieur aux filtres conventionnellement utilisées en polarimétrie.Les résultats des modèles les plus profonds obtiennent les meilleures performances pour tous les indicateurs sur l’ensemble des données homogènes simulées. Dans le cas des données en patchwork, les résultats pour la restauration des détails sont nettement favorables au filtrage par RNC les plus profonds.L’application du filtrage par RNC sur les images satellitaires RADARSAT-2,ALOS/PalSAR ainsi GaoFen-3 montre des résultats comparables ou supérieurs aux filtres conventionnels. Les meilleurs résultats ont été obtenus par le modèle à 5 couches cachées(si on ne compte pas la couche d’entrée et de sortie), avec 8 filtres 3×3 par couche convolutive, sauf pour la couche d’entrée où la taille des filtres étaient de 9×9. Par contre,les données d’apprentissage doivent être bien ajustées à l’étendue des statistiques des images polarimétriques réelles pour obtenir de bon résultats. Ceci est surtout vrai au niveau de la modélisation des cibles ponctuelles dont la restauration semblent plus difficiles. / Due to the coherent nature of the Synthetic Aperture Radar (SAR) signal, polarimetric SAR(POLSAR) images are affected by speckle noise. The effect of speckle can be so severe as to render the POLSAR data unusable. This is especially true for single-look data that suffer from very intense speckle. Noise filtering is necessary to improve the estimation of polarimetric parameters that can be computed from this type of data. This is an important step in the processing and analysis of POLSAR images. Recently, a new approach has emerged in data processing aimed at solving a multi-tude of problems including filtering, image restoration, speech recognition, classification orimage segmentation. This approach is deep learning and convolutional neural networks(CONVNET). Recent works show that CONVNET are a promising alternative for filtering SAR images. Indeed, by their ability to learn an optimal filtering model only from the data, they tend to outperform classical approaches to filtering on SAR images. The objective of this study is to analyze and evaluate the effectiveness of CONVNET filtering on simulated POLSAR data and on RADARSAT-2, ALOS/PalSAR and GaoFen-3 satellite images acquired over the San Francisco urban area (California). Models inspired by the architecture of a CONVNET used in particular in super-resolution have been adapted for the filtering of the polarimetric coherency matrix. The effect of different structural parameters of theCONVNET architecture on filtering were analyzed, among which are the depth of the neural network (the number of stacked layers), the width of the neural network (the number of filters per convoluted layer) and the size of the filters of the first convolution layer. The models were learned by backpropagation of the error gradient using 3 datasets that simulate single-look polarimetry of the scatterers according to Cloude-Pottier classes. The first dataset contains only homogeneous areas. The last two datasets consist of patchwork simulations where local intensity is simulated by texture images and point target are added to the patchwork in the case of the last dataset. The performance of the different filters by CONVNET was measured by indicators including relative error on the estimation of polarimetric signatures and decomposition parameters as well as distortion measurements on the recovery of major details and on the conservation of point targets.The results show that CONVNET filtering of polarimetric data is either equivalent or significantly superior to conventional polarimetric filters. The results of the deepest models obtain the best performance for all indicators over the simulated homogeneous dataset. Inthe case of patchwork dataset, the results for detail restoration are clearly favourable to the deepest CONVNET filtering. The application of CONVNET filtering on RADARSAT-2, ALOS/PalSAR andGaoFen-3 satellite images shows results comparable or superior to conventional filters. The best results were obtained by the 5 hidden layers model (not counting the input and outputlayers), with 8 filters 3×3 per convolutional layer, except for the input layer where the filtersize was 9×9. On the other hand, the training data must be well adjusted to the statistical range of the real polarimetric images to obtain good results. This is especially true when modeling point targets that appear to be more difficult to restore.