Дисертації з теми "Images à résolution submétrique"
Оформте джерело за APA, MLA, Chicago, Harvard та іншими стилями
Ознайомтеся з топ-50 дисертацій для дослідження на тему "Images à résolution submétrique".
Біля кожної праці в переліку літератури доступна кнопка «Додати до бібліографії». Скористайтеся нею – і ми автоматично оформимо бібліографічне посилання на обрану працю в потрібному вам стилі цитування: APA, MLA, «Гарвард», «Чикаго», «Ванкувер» тощо.
Також ви можете завантажити повний текст наукової публікації у форматі «.pdf» та прочитати онлайн анотацію до роботи, якщо відповідні параметри наявні в метаданих.
Переглядайте дисертації для різних дисциплін та оформлюйте правильно вашу бібліографію.
Li, Sizhuo. "Deep Learning for Forest Resource Mapping from Sub-Meter Resolution Imagery : Technical Insights and Methodologies." Electronic Thesis or Diss., université Paris-Saclay, 2024. http://www.theses.fr/2024UPASJ015.
Повний текст джерелаDeep learning has transformed numerous fields so far, propelled by improved algorithms and increased data accessibility. Remote sensing, in particular, offers significant opportunities for vision applications with direct socio-ecological implications. Forests represent a major environmental component, offering essential functions including climate regulation, biodiversity preservation, and interaction with living creatures. Departing from studies that reveal forest patterns using coarse resolution imagery or structural sensors like LiDAR, this thesis explores sub-meter resolution imagery - human-interpretable and highly detailed visual data covering the Earth. Forest attributes of different types are investigated, varying from tree characteristics to forest height and biomass. Technically, this thesis starts with a vanilla setup of in-domain semantic segmentation, delves into the regression of structural attributes, and ends with cross-domain adaptation. This brings insights into the learning capacity of vision models in the context of natural scene image understanding. The first part of the thesis introduces a deep learning framework to count, locate, and estimate the height of individual trees from aerial images at the national scale. An attention UNet is utilized to delineate individual tree crowns and count trees with point supervision. Tree height is estimated from optical imagery by learning a mapping from visual cues to canopy heights projected from LiDAR point clouds. Results are compared against field data to assess the practical values of the framework as supplementary data to support digitized national forest management. This study highlights the capacity of deep learning in characterizing visually interpretable forest and tree structures. Yet, challenges persist in learning more complex patterns from the optical data. This motivates the second study on forest biomass at stand level, a structural measure of forests typically collected on the ground. At larger scales, predominating methods apply statistical or machine learning models on multispectral imagery, often complemented by height data and calibrated with field data. To our knowledge, we are the first to demonstrate that stand-level biomass can be directly learned from sub-meter resolution RGB imagery, rich in forest and tree details, using convolutional neural networks and field data. This provides avenues for efficient and accurate quantification of forest biomass, a critical indicator of forest resources that supports nature preservation and carbon-neutral commitments. The first two studies employ deep learning systems to quantify forest attributes given a specific dataset, which follows the principal in-domain assumption of machine learning. Yet, degraded performance is often observed when applied to out-of-distribution data, a common scenario in practice. The third study aims to address the domain shift issue, exploring whether deep learning models trained on one dataset can be quickly adapted to new datasets with marginal efforts. We release a new dataset consisting of sub-meter resolution optical imagery collected in five countries and assess the cross-domain adaptability of various image-level regression tasks, including tree cover, total tree count, and average canopy height. By enforcing ordered embedding space during training, models are effectively prepared for later adaptation in source-free low-shot setups. Overall, this thesis introduces a collection of deep learning systems tailored for forest resource mapping with depth into technical and applied insights, contributing to sustainable management efforts for a greener future
Deep learning har hidtil transformeret talrige områder, drevet af forbedrede algoritmer og øgettilgængelighed af data. Fjernregistrering tilbyder betydelige muligheder for visionsapplikationermed direkte socioøkologiske implikationer. Skoveudgør en vigtig miljømæssig komponent og tilbyder essentielle funktioner, herunder klimaregulering, bevarelse af biodiversitet og interaktion medlevende væsener. Afgang fra studier, der afslørerskovmønstre ved hjælp af grovopløsningsbilledereller struktursensorer som LiDAR, udforsker denneafhandling sub-meteropløsningsbilleder - menneskefortolkelige og detaljerede visuelle data, derdækker Jorden. Forskellige skovattributter undersøges, lige fra trækarakteristika til skovhøjdeog biomasse. Teknisk set starter denne afhandling med en grundlæggende opsætning af semantisk segmentering inden for domænet, går videretil regression af strukturelle attributter og sluttermed tværfaglig tilpasning. Dette giver indsigt i visionmodellers indlæringskapacitet i konteksten afforståelse af naturscener. Afhandlingens førstedel introducerer et dybtlæringsrammeværk til attælle, lokalisere og estimere højden af individuelle træer fra luftfotos på nationalt plan. En opmærksomhed UNet bruges til at afgrænse individuelle trækroner og tælle træer med punktsupervision. Træhøjde estimeres fra optisk billedmaterialeved at lære en afbildning fra visuelle ledetråde tilkronhøjder projiceret fra LiDAR-punktskyer. Resultater sammenlignes med feltdatabaser for at vurdere rammeværkets praktiske værdier som supplement til understøttelse af digitaliseret nationalskovforvaltning. Denne undersøgelse fremhæverdybtlæringens evne til at karakterisere visuelt fortolkede skov- og træstrukturer. Dog vedvarerudfordringer i at lære mere komplekse mønstrefra det optiske data. Dette motiverer den anden undersøgelse af skovbiomasse på bestandsniveau, en strukturel måling af skove, der typiskindsamles på jorden. På større skalaer anvender dominerende metoder statistiske eller maskinlæringsmodeller på multispektralt billedmateriale,ofte suppleret med højdedata og kalibreret medfeltdatabaser. Efter vores viden er vi de førstetil at demonstrere, at bestandsniveauets biomassedirekte kan læres fra sub-meteropløsnings-RGBbilledmateriale, rigt på skov- og trædetaljer, vedhjælp af konvolutionelle neurale netværk og feltdatabaser. Dette giver muligheder for effektiv og præcis kvantificering af skovbiomasse, enkritisk indikator for skovressourcer, der støtternaturbevarelse og klimaneutrale forpligtelser. Deførste to studier benytter dybtlæringssystemer tilat kvantificere skovattributter ud fra et specifikt datasæt, hvilket følger den grundlæggendeantagelse om maskinlæring inden for domænet.Dog observeres der ofte nedsat præstation, nårdet anvendes på data uden for distributionen,en almindelig situation i praksis. Det tredjestudie sigter mod at adressere domæneskiftproblemet ved at undersøge, om dybtlæringsmodeller trænet på ét datasæt hurtigt kan tilpassestil nye datasæt med marginale bestræbelser.Vi frigiver et nyt datasæt bestående af submeteropløsnings optisk billedmateriale indsamleti fem lande og vurderer krydsdomænet tilpasningsevne for forskellige billedniveau-regressionstasks,herunder trædække, samlet trætælling og gennemsnitlig krones højde. Ved at håndhæve en ordnetindlejret rum under træning forberedes modellereffektivt til senere tilpasning i kildefrie lavskudsopsætninger. Overordnet introducerer denne afhandling en samling dybtlæringssystemer skræddersyet til kortlægning af skovressourcer med dybdei tekniske og anvendte indsigter, hvilket bidrager tilbæredygtige forvaltningsindsatser for en grønnerefremtid
Ploquin, Marie. "Super résolution pour l'amélioration de la résolution des images échographiques." Thesis, Tours, 2011. http://www.theses.fr/2011TOUR4025/document.
Повний текст джерелаMedical Imaging Ultrasound has several advantages such as its safety, ease of use, the diversity of organs that can be imaged and the low cost of this imaging mode. However, the images obtained by ultrasound suffer from relatively low resolution compared to others than can be obtain with an MRI or using X-rays. The major challenge of medical ultrasound is the ability to produce images with a resolution much finer, without modifying the nominal frequency.Work has been undertaken in this direction for some time. Several approaches have been explored. Most of the work done so far has been to work on the ultrasound acquiring device and particularly on ultrasonic probes, with main objective to increase the frequency of ultrasound used. This approach has led to the existence of high-resolution ultrasound, but with the reduction of the depth of exploration as an important limitation.Another approach is to treat numerically conventional ultrasound images to improve resolution. This method has several advantages, it allows to circumvent such difficulties caused by the reduction of depth of field due to the increase in the ultrasonic frequency.In this thesis, we present a method to improve the resolution of ultrasound images. The thesis to be to adapt this method to ultrasound imaging and to provide an estimate of the maximum theoretical resolution achieved by this method based on image parameters including SNR and the bandwidth of the PSF. We also proposed a method of superresolution suitable for ultrasound. By providing on improving theoretical superresolution and adaptation to the particular case of ultrasound, this thesis opens up on improving the resolution of ultrasound images by processing the signal and the image
Bouzkraoui, Mohammed. "Détection des arbres individuels dans des images de haute résolution." Thesis, National Library of Canada = Bibliothèque nationale du Canada, 1999. http://www.collectionscanada.ca/obj/s4/f2/dsk1/tape7/PQDD_0021/MQ47178.pdf.
Повний текст джерелаErus, Güray. "Reconnaissance d'objets cartographiques dans les images satellitaires à haute résolution." Paris 5, 2008. http://www.theses.fr/2008PA05S003.
Повний текст джерелаOur thesis' subject considers specifically the recognition of cartographic objects with a highly composite structure, such as roundabouts and bridges, on high resolution satellite images. Being man-made constructions, cartographic objects have regular geometrical features that distinguish them from other objects. We propose to exploit principally these features in order to obtain a representation of their inherent structure. A method to generate an explicit structural object model represented by Attributed Relational Graphs (ARGs) from images segmented by an expert is first developed. At the end of this preliminary stage we succeeded to generate object models for the roundabout and bridge categories. We then proposed to learn a more flexible model based on the appearances of the local parts of the objects using a statistical learning method. An implicit object model is learned by the fusion of weak classifiers obtained from geometrical primitives using the Adaboost algorithm. An object recognition method using an implicit model constructed by a parts and structures approach is then proposed. The model is learned using the Mean-Shift clustering algorithm. Finally, the methods are validated on satellite images provided by the CNES in the frame of a national competition and a cartographic application
Malek, Mohamed. "Extension de l'analyse multi-résolution aux images couleurs par transformées sur graphes." Thesis, Poitiers, 2015. http://www.theses.fr/2015POIT2304/document.
Повний текст джерелаIn our work, we studied the extension of the multi-resolution analysis for color images by using transforms on graphs. In this context, we deployed three different strategies of analysis. Our first approach consists of computing the graph of an image using the psychovisual information and analyzing it by using the spectral graph wavelet transform. We thus have defined a wavelet transform based on a graph with perceptual information by using the CIELab color distance. Results in image restoration highlight the interest of the appropriate use of color information. In the second strategy, we propose a novel recovery algorithm for image inpainting represented in the graph domain. Motivated by the efficiency of the wavelet regularization schemes and the success of the nonlocal means methods we construct an algorithm based on the recovery of information in the graph wavelet domain. At each step the damaged structure are estimated by computing the non local graph then we apply the graph wavelet regularization model using the SGWT coefficient. The results are very encouraging and highlight the use of the perceptual informations. In the last strategy, we propose a new approach of decomposition for signals defined on a complete graphs. This method is based on the exploitation of of the laplacian matrix proprieties of the complete graph. In the context of image processing, the use of the color distance is essential to identify the specificities of the color image. This approach opens new perspectives for an in-depth study of its behavior
Corpetti, Thomas. "Images & télédétection : analyse de séquences à basse et très haute résolution spatiale." Habilitation à diriger des recherches, Université Rennes 1, 2011. http://tel.archives-ouvertes.fr/tel-00616558.
Повний текст джерелаLehureau, Gabrielle. "Fusion de données optique et radar à haute résolution en milieu urbain." Paris, Télécom ParisTech, 2010. http://www.theses.fr/2010ENST0035.
Повний текст джерелаThe increasing quality of satellite images has generated interest in extracting man-made structures. Optical and radar sensors deliver images with unlike physical properties, thus it is interesting to fuse such images in order to benefit from joint observation. Such a process begins with registration of these images. We propose an automatic registration of radar and optical images without using sensors parameters. First, a rigid transformation is determined using a multi-scale pyramid of features representing the contours of roads and buildings. Secondly, a polynomial transformation is determined. The coefficients are obtained by associating points in both images using mutual information. We also developped a classification process in order to identify all scene objects. This method used both information from optical and radar images and svm classifier. We proved in this part a good robustness to segmentation and the interest of using both data to improve the classification, especially for roads and buildings. Finally we present an original method of fine registration for the buildings based on the assumption that “a trained classifier can recognize registrated buildings from unregistrated“. So, buildings are classified considering many translations in order to determine the good one. We also show the importance of contextual information to improve the fine registration, especially for buildings
Puissant, Anne. "INFORMATION GÉOGRAPHIQUE ET IMAGES A TRÈS HAUTE RÉSOLUTION UTILITÉ ET APPLICATIONS EN MILIEU URBAIN." Phd thesis, Université Louis Pasteur - Strasbourg I, 2003. http://tel.archives-ouvertes.fr/tel-00467474.
Повний текст джерелаSimonetto, Elisabeth. "Extraction 3-D de structures industrielles sur des images ramsès haute résolution par radargrammétrie." Rennes 1, 2002. https://tel.archives-ouvertes.fr/tel-00749513.
Повний текст джерелаMees, Wim. "Contribution à l'analyse distribuée de scènes : application aux images satellitaires multi spectrales, haute résolution." Nancy 1, 2000. http://www.theses.fr/2000NAN10282.
Повний текст джерелаPuissant, Anne. "Information géographique et images à très haute résolution : Utilité et applications en milieu urbain." Université Louis Pasteur (Strasbourg) (1971-2008), 2003. https://publication-theses.unistra.fr/public/theses_doctorat/2003/PUISSANT_Anne_2003.pdf.
Повний текст джерелаSince the mid 1990s 'INFO-STRATEGY', i. E. The strategic use of Geographic Information, is one of the reliable goals in urban management, planning and sustainable development. Geographic Information requires multisource and multiscale data, updated on a regular basis at appropriate time periods. Earth Observation data-such as aerial photography and satellite image-represent an important source for GI. Recently, the use of Very High Resolution imagery data (VHR - finer than 5m) offers an abundance of numerical information and therefore an important possibility to use GI in decision-making concerning the urban environment. A reflection is thus required in terms of needs rather than supply in order to meet the requirements of localised information on a variety of scales. Moreover, defining the needs of the 'end users' (decision-makers, managers, technicians) corresponds to propose the necessary adjustment to the definition of the capacity of these new sensors. In this context, on the basis of comparisons and surveys of the 'end users' a reference grid defining the needs of GI on a large-scale basis has been established. Some "potential applications" of metric images are highlighted. Tests have been carried out in order to analyse very precisely the benefits of use of these new sensors. An important factor in the increase of spatial resolution is a new vision of the territory that is closer to real-ity. Urban objects are individually recognisable (field of identification) and can be characterised by their components (field of analysis). This superabundance of details disturbs the classical automatic proce-dures of extraction (per pixel classification) and makes complex the attribution of objects to existing nomenclatures. After having tested several traditional algorithms used on HR images, it appears neces-sary to transcribe the rules of identification of urban objects and to integrate them into an 'object-oriented' classification method
Lauer-Leredde, Christine. "Analyse haute-résolution des sédiments non-consolidés : mesures géophysiques, images électriques, minéralogie et modélisation." Aix-Marseille 2, 1997. http://www.theses.fr/1997AIX22129.
Повний текст джерелаFallourd, Renaud. "Suivi des glaciers alpins par combinaison d'informations hétérogènes : images SAR Haute Résolution et mesures terrain." Phd thesis, Université de Grenoble, 2012. http://tel.archives-ouvertes.fr/tel-00718596.
Повний текст джерелаDevaux, Nicolas. "Images très haute résolution et analyse spatiale pour localiser l'habitat rural des pays en développement." Orléans, 2006. http://www.theses.fr/2006ORLE1075.
Повний текст джерелаWithin the context of developing southern countries rural areas, deployment of electrification sometime relies on decentralized, rural electrification projects manage by foreign companies. The realization and management of such projects are faced with a lack of high scale cartographic data, for individual localisation of rural houses to be electrified. This thesis demonstrates that high resolution ; satellites images make a pertinent data source to partially fill this cartographic shortage, thanks to a better adequacy of their spatial and spectral characteristics with the physical specificities of the studied houses. Nevertheless, identification and localisation of houses are strongly disrupted by their dilution in an environment where building materials remain mainly natural in origin. Consequently, the systematic procedures of houses recognition imply to overcome known limits of traditional remote sensing treatments. It is, therefore, necessary to give a better consideration for houses spatial and functional contextualisation within rural areas. Thus, we are proposing a subdivision of the rural areas in geographic primitives of interest, combined to the study of their spatial links with houses. Results are integrated within multilevel images treatments where more easily identifiable geographic primitives facilate houses localisation. According to this method, on object-oriented approach and another one, built up thanks to image analysis, were tested on an investigation field of Extreme North Cameroon. Actual results are promising, but not sufficient for a direct applicative transfer. Nevertheless, they do demonstrate the benefit and the need to elaborate treatments from the modelling of rural habitations spatial context, which requires putting an emphasis on spatial analysis
Aubert, Didier. "Mise en correspondance d'indices images en résolutions multiples." Phd thesis, Grenoble INPG, 1989. http://tel.archives-ouvertes.fr/tel-00332349.
Повний текст джерелаCantalloube, Faustine. "Détection et caractérisation d'exoplanètes dans des images à grand contraste par la résolution de problème inverse." Thesis, Université Grenoble Alpes (ComUE), 2016. http://www.theses.fr/2016GREAY017/document.
Повний текст джерелаDirect imaging of exoplanets provides valuable information about the light they emit, their interactions with their host star environment and their nature. In order to image such objects, advanced data processing tools adapted to the instrument are needed. In particular, the presence of quasi-static speckles in the images, due to optical aberrations distorting the light from the observed star, prevents planetary signals from being distinguished. In this thesis, I present two innovative image processing methods, both based on an inverse problem approach, enabling the disentanglement of the quasi-static speckles from the planetary signals. My work consisted of improving these two algorithms in order to be able to process on-sky images.The first one, called ANDROMEDA, is an algorithm dedicated to point source detection and characterization via a maximum likelihood approach. ANDROMEDA makes use of the temporal diversity provided by the image field rotation during the observation, to recognize the deterministic signature of a rotating companion over the stellar halo. From application of the original version on real data, I have proposed and qualified improvements in order to deal with the non-stable large scale structures due to the adaptative optics residuals and with the remaining level of correlated noise in the data. Once ANDROMEDA became operational on real data, I analyzed its performance and its sensitivity to the user-parameters proving the robustness of the algorithm. I also conducted a detailed comparison to the other algorithms widely used by the exoplanet imaging community today showing that ANDROMEDA is a competitive method with practical advantages. In particular, it is the only method that allows a fully unsupervised detection. By the numerous tests performed on different data set, ANDROMEDA proved its reliability and efficiency to extract companions in a rapid and systematic way (with only one user parameter to be tuned). From these applications, I identified several perspectives whose implementation could significantly improve the performance of the pipeline.The second algorithm, called MEDUSAE, consists in jointly estimating the aberrations (responsible for the speckle field) and the circumstellar objects by relying on a coronagraphic image formation model. MEDUSAE exploits the spectral diversity provided by multispectral data. In order to In order to refine the inversion strategy and probe the most critical parameters, I applied MEDUSAE on a simulated data set generated with the model used in the inversion. To investigate further the impact of the discrepancy between the image model used and the real images, I applied the method on realistic simulated images. At last, I applied MEDUSAE on real data and from the preliminary results obtained, I identified the important input required by the method and proposed leads that could be followed to make this algorithm operational to process on-sky data
Nabucet, Jean. "Apport des données de télédétection à très haute résolution spatiale pour la cartographie de la végétation en milieu urbain." Thesis, Rennes 2, 2018. http://www.theses.fr/2018REN20062/document.
Повний текст джерелаAbstract: Detailed knowledge and monitoring of urban vegetation is an important issue, both for scientists studying landscape-ecosystems relationships, and for the managers who are in charge of the vegetation management. The main objective of this thesis is to evaluate the interest of THRS images to map urban vegetation. More specifically, it aims to evaluate the potential of THRS images on three components: spatial resolution, spectral resolution and altimetry. For this purpose, we processed several types of THRS optical data acquired on the City of Rennes and the Prairies Saint-Martin Site: multispectral and superspectral 2D images, 2.5D multispectral images and 3D data acquired with a bi-spectral LiDAR. Firstly, we assessed the interest of using 2D multispectral THRS images to identify and characterize vegetation and superspectral THRS images to discriminate plant species. Secondly, we assessed the contribution of THRS 2.5D and 3D multispectral data to map vegetation patterns in urban areas using spectral, contextual and height variables. Thirdly, we sought to evaluate the impact, of the integration of vegetation information derived from THRS remote sensing data into two environmental models, one to study the landscape-biodiversity relationship, the other to analyze the landscape- urban cool island relationship
Dusseux, Pauline. "Exploitation de séries temporelles d'images satellites à haute résolution spatiale pour le suivi des prairies en milieu agricole." Thesis, Rennes 2, 2014. http://www.theses.fr/2014REN20031/document.
Повний текст джерелаIn agricultural areas, we observed a decrease of grasslands and change in their management in the last half–century, which are commonly associated with agriculture intensification. These changes have affected environmental and economic systems. In this context, the evaluation of grassland status and grassland management in farming systems is a key–issue for sustainable agriculture. With the arrival of new Earth observation sensors with high spatial and temporal resolutions, it is now possible to study grasslands at fine scale using regular observations over time. The objective of this thesis is to identify grasslands and their management practices using parameters derived from time–series of high spatial resolution (HSR) remote sensing data. For that purpose, several intra–annual time series of HSR optical and Synthetic Aperture Radar (SAR) satellite images were acquired in order to identify grasslands and three of their management practices: grazing, mowing and mixed management, on a catchment area mainly oriented towards cattle production. Results obtained from the processing and analysis of the optical time series have shown that it is possible to estimate with good accuracy grassland biomass, to identify and to characterize them. They also highlighted that radar images improve grassland identification without being able to distinguish management practices, the combined use of the two types of images further increasing grassland identification. Furthermore, results showed that the classification methods based on comparison criteria adapted to time series (warping criteria) increase significantly results for discriminating grassland management practices compared to standard classification methods
Chesnel, Anne-Lise. "Quantification de dégâts sur le bâti liés aux catastrophes majeures par images satellite multimodales très haute résolution." Phd thesis, École Nationale Supérieure des Mines de Paris, 2008. http://pastel.archives-ouvertes.fr/pastel-00004211.
Повний текст джерелаLavergne, Jean-Louis. "Imagerie chimique à haute résolution spatiale "images-spectres" : Application à la caractérisation des cristaux composites d'halogénures d'argent." Ecully, Ecole centrale de Lyon, 1994. http://www.theses.fr/1994ECDL0036.
Повний текст джерелаBenelcadi, Hajar. "Apport de l’analyse texturale des images radar à haute résolution spatiale pour la cartographie des forêts tropicales." Thesis, Paris Est, 2014. http://www.theses.fr/2014PEST1161/document.
Повний текст джерелаSince 2007, a new generation of SAR sensors (Synthetic Aperture RADAR) was launched. These sensors (TerraSAR-X, Cosmo-SkyMed, RADARSAT-2, and Sentinel) are characterized by metric spatial resolutions unlike previous sensors (ERS, JERS, ALOS, ASAR) with a spatial resolution of about twenty meters. Metric spatial resolution highlights interesting textural information that was inaccessible with the previously existing SAR sensors. This thesis aims at evaluating the contribution of textural analysis from high spatial resolution images for tropical forests mapping. Three different study sites with different problematic have been chosen to evaluate the textural analysis in Cambodia, Cameroun and Brazil. Indeed, the contribution of the analysis of textural information for classification has been emphasized. The latter is understood through the analysis of Haralick textural parameters, second order statistic parameters. The retained algorithm of classification is the SVM (Support Vector Machine), as it allows taking into account numerous parameters, which can be heterogeneous with respect to their physical dimension
Dellière, Julien. "Simulation de données SAR en milieu urbain à haute résolution : étude de faisabilité d'une méthode exacte." Paris, ENST, 2008. http://www.theses.fr/2008ENST0013.
Повний текст джерелаThe data that will be received from the future Earth'observation system will have a resolution about 1 m. Without being comparable with the resolutions of the air sensors like Ramses, this one is clearly better than the resolutions reached by the existing satellite sensors like ERS, RadarSat or EnviSat. This evolution, that opens the field of many applications like DEM (Digital Elevation Model) reconstruction, makes difficult a good understanding of the radar signal' formation to high and very high resolution (THR) and in urtan environment. Indeed, the backscattering mechanisms and the matter/ radiation interactions are still badly controlled in urban environment and when the pixel cannot be considered any more as having great dimension compared to the wavelength. In fact all attempt to understand the high resolution radar signal prove to be abortive today because our knowledge of the matter/ radiation interaction processes on very complex surfaces as the city is very insufficient. The purpose of this work is to go back to the origins of the electromagnetism processes in order to build a simulator able to help to understand the radar images. We feel that going into details of the radar images formation processes for this kind of resolution and for this kind of matter should make it possible to better understand the potentialities and the limits of the SAR imagery and interferometry
Loncan, Laëtitia. "Fusion of hyperspectral and panchromatic images with very high spatial resolution." Thesis, Université Grenoble Alpes (ComUE), 2016. http://www.theses.fr/2016GREAT065/document.
Повний текст джерелаStandard pansharpening aims at fusing a panchromatic image with a multispectral image in order to synthesize an image with the high spatial resolution of the former and the spectral resolution of the latter. In the last decade many pansharpening algorithms have been presented in the literature using multispectral data. With the increasing availability of hyperspectral systems, these methods are now extending to hyperspectral pansharpening, i.e. the fusion of a panchromatic image with a high spatial resolution and a hyperspectral image with a coarser spatial resolution. However, state of the art hyperspectral pansharpening methods usually do not consider the problem of the mixed pixels. Their goal is solely to preserve the spectral information while adding spatial information. In this thesis, in a first part, we present the state-of-the-art methods and analysed them to identified there performances and limitations. In a second part, we present an approach to actually deal with mixed pixels as a pre-processing step before performing the fusion. This improves the result by adding missing spectral information that is not directly available in the hyperspectral image because of the mixed pixels. The performances of our proposed approach are assessed on different real data sets, with different spectral and spatial resolutions and corresponding to different contexts. They are compared qualitatively and quantitatively with state of the art methods, both at a global and a local scale
Péteri, Renaud. "Extraction de réseaux de rues en milieu urbain à partir d' images satéllites à très haute résolution spatiale." Paris, ENMP, 2003. http://www.theses.fr/2003ENMP1141.
Повний текст джерелаThis report proposes a method for extracting urban street networks from new very high spatial resolution satellite images. Its goal is to meet the need for an automatized building of maps. The proposed method uses digital image as only input data. It is semi-automatic at the detection step, and takes advantage of cooperation between linear representation of streets and their representation as surface elements. A topological graph of the street network is first extracted, and used for initializing the surface reconstruction step. The extraction result can then be used in order to precisely register the street centerline. This method favors strong geometrical constraints in order to avoid a radiometric profile model of the street, too variable in urban areas. To that aim, a model of active contour associated with the wavelet transform, called doublesnake, has been developed. Its evolution in a multi-scale framework enables the extraction of parallel street sides in a noisy environment. Then, final positions of doublessnakes permit the extraction of intersections. The method has been applied on images from different sensors and with different urban types. An innovative protocol for a quantitative assessment of the results compared to human interpretation has shown its generic aspect, as well as its robustness with respect to noise. This method is a step toward a fully automatized cartography of the street network
Le, Men Camille. "Segmentation Spatio-temporelle d'une séquence d'images satellitaires à haute résolution." Phd thesis, Ecole nationale supérieure des telecommunications - ENST, 2009. http://pastel.archives-ouvertes.fr/pastel-00658159.
Повний текст джерелаRouvière, Clémentine. "Experimental parameter estimation in incoherent images via spatial-mode demultiplexing." Electronic Thesis or Diss., Sorbonne université, 2024. http://www.theses.fr/2024SORUS033.
Повний текст джерелаHistorically, the resolution of optical imaging systems was dictated by diffraction, and the Rayleigh criterion was long considered an unsurpassable limit. In superresolution microscopy, this limit is overcome by manipulating the emission properties of the object. However, in passive imaging, when sources are uncontrolled, reaching sub-Rayleigh resolution remains a challenge. Here, we implement a quantum-metrology-inspired approach for estimating the separation between two incoherent sources, achieving a sensitivity five orders of magnitude beyond the Rayleigh limit. Using a spatial mode demultiplexer, we examine scenes with bright and faint sources, through intensity measurements in the Hermite-Gauss basis. Analysing sensitivity and accuracy over an extensive range of separations, we demonstrate the remarkable effectiveness of demultiplexing for sub-Rayleigh separation estimation. These results effectively render the Rayleigh limit obsolete for passive imaging
Stankov, Katia. "Détection des bâtiments à partir des images multispectrales à très haute résolution spatiale par la transformation Hit-or-Miss." Thèse, Université de Sherbrooke, 2014. http://savoirs.usherbrooke.ca/handle/11143/108.
Повний текст джерелаAuclair-Fortier, Marie-Flavie. "Résolution du problème de transfert de chaleur par une approche TAC : application au traitement et à l'analyse des images." Thèse, Université de Sherbrooke, 2004. http://savoirs.usherbrooke.ca/handle/11143/5029.
Повний текст джерелаBirjandi, Payam. "Modélisation et extraction des descripteurs intrinsèques des images satellite à haute résolution : approches fondées sur l'analyse en composantes indépendantes." Phd thesis, Paris, Télécom ParisTech, 2011. https://pastel.hal.science/pastel-00677956.
Повний текст джерелаSub-meter resolution satellite images, capture very detailed information, as for example, shape of buildings, roads, etc. The main purpose of the thesis is to propose descriptors for sub-meter resolution satellite images especially for those who contain geometrical or man-made structures. Independent component analysis (ICA) is a good candidate for this purpose, since previous studies demonstrated that the resulted basis vectors contain some small lines and edges, the important elements in the characterization of geometrical structures. As a basic analysis, a study about the effects of scale size and dimensionality of ICA system on indexing of satellite images is presented and the optimum dimensionality and scale size are found. There are two view points for feature extraction based on ICA. The usual idea is to use the ICA coefficients (ICA sources) and the other is to use the ICA basis vectors related to every image. Based on the first point of view, an ordinary ICA source based approach is proposed for feature extraction. This approach is developed and modified through a topographic ICA system to extract middle level features which leads to a significant improvement in results. Based on other point of view, two methods are proposed. One of them uses the bag of words idea which considers the basis vectors as visual words. Second method uses the lines properties inside the basis vectors to extract features. Also, using the lines properties idea, another method is developed which directly detects the line segments in the images. Finally, the capabilities of proposed descriptors are compared through a supervised classification based on support vector machine (SVM)
Birjandi, Payam. "Modélisation et Extraction des Descripteurs Intrinsèques des Images Satellite à Haute Résolution: Approches Fondées sur l'Analyse en Composantes Indépendantes." Phd thesis, Télécom ParisTech, 2011. http://pastel.archives-ouvertes.fr/pastel-00677956.
Повний текст джерелаZhao, Ningning. "Inverse problems in medical ultrasound images - applications to image deconvolution, segmentation and super-resolution." Phd thesis, Toulouse, INPT, 2016. http://oatao.univ-toulouse.fr/16613/1/Zhao.pdf.
Повний текст джерелаJeanrot-Gouzy, Nicole. "Ontogenèse normale et pathologique de la fonction visuelle chez l'enfant : pouvoir de résolution et vision binoculaire." Toulouse 3, 1991. http://www.theses.fr/1991TOU30198.
Повний текст джерелаLavoie, André. "Potentiel des images satellitaires multibandes à haute résolution spatiale pour la cartographie des componsants de l'eau en milieu côtier marin." Thesis, National Library of Canada = Bibliothèque nationale du Canada, 1997. http://www.collectionscanada.ca/obj/s4/f2/dsk2/ftp03/NQ26803.pdf.
Повний текст джерелаAmani, Massalabi. "Détection et exploitation d'ombre de bâti sur les images de très haute résolution spatiale (IKONOS) application au milieu urbain (Sherbrooke)." Thèse, Université de Sherbrooke, 2006. http://savoirs.usherbrooke.ca/handle/11143/2751.
Повний текст джерелаDupuis, Olivier. "Fusion entre les données ultrasonores et les images de radioscopie à haute résolution : application au contrôle de cordon de soudure." Lyon, INSA, 2000. http://theses.insa-lyon.fr/publication/2000ISAL0093/these.pdf.
Повний текст джерелаThis thesis focuses on the development of radioscopic (XR) and ultrasonic (US) data fusion for the automatic inspection of steel welded joints in a way to enhance the reliability of defect detection. The mathematical model is the theory of evidence of Dempster-Shafer. The study of physical laws leading to the formation of XR image and US signal helped for the development of a specific processing for the detection and matching of defects. Unfortunately, the detection of low amplitude signal defects also yields false alarm detection. We therefore developed a training stage to attribute a confidence level to a detected object. During this stage, different features of reference defects were calculated (contrast-to-noise ratio, area, elongation…) and compared to the interpretation of human expert analysis. We distinguish different areas of the features space in which some types of objects are predominant. A novel method has been developed for attributing a degree of belief to an unknown object taking both uncertainty and imprecision into account. Eventually, the data fusion stage consists in combining confidence levels to increase the confidence in the presence of a defect, but also to precise its nature and dimensions
Rousselle, Denis. "Classification d’objets au moyen de machines à vecteurs supports dans les images de sonar de haute résolution du fond marin." Thesis, Rouen, INSA, 2016. http://www.theses.fr/2016ISAM0020.
Повний текст джерелаThis thesis aims to improve the classification of underwater objects in high resolution sonar images. Especially, we seek to make the distinction between mines and harmless objects from a collection of mine-like objects. Our research was led by two classical constraints of the mine warfare : firstly, the lack of data and secondly, the need for readability of the classification. In this context, we built a database as much representative as possible and simulated objects in order to complete it. The lack of examples led us to use a compact representation, originally used by the face recognition community : the Structural Binary Gradient Patterns (SBGP). To the same end, we derived a method of semi-supervised domain adaptation, based on optimal transport, that can be easily interpreted. Finally, we developed a new classification algorithm : the Ensemble of Exemplar-Maximum Excluding Ball (EE-MEB) which is suitable for small datasets and with an easily interpretable decision function
Rousselle, Denis. "Classification d’objets au moyen de machines à vecteurs supports dans les images de sonar de haute résolution du fond marin." Electronic Thesis or Diss., Rouen, INSA, 2016. http://www.theses.fr/2016ISAM0020.
Повний текст джерелаThis thesis aims to improve the classification of underwater objects in high resolution sonar images. Especially, we seek to make the distinction between mines and harmless objects from a collection of mine-like objects. Our research was led by two classical constraints of the mine warfare : firstly, the lack of data and secondly, the need for readability of the classification. In this context, we built a database as much representative as possible and simulated objects in order to complete it. The lack of examples led us to use a compact representation, originally used by the face recognition community : the Structural Binary Gradient Patterns (SBGP). To the same end, we derived a method of semi-supervised domain adaptation, based on optimal transport, that can be easily interpreted. Finally, we developed a new classification algorithm : the Ensemble of Exemplar-Maximum Excluding Ball (EE-MEB) which is suitable for small datasets and with an easily interpretable decision function
Briand, Thibaud. "Image Formation from a Large Sequence of RAW Images : performance and accuracy." Thesis, Paris Est, 2018. http://www.theses.fr/2018PESC1017/document.
Повний текст джерелаThe aim of this thesis is to build a high-quality color image, containing a low level of noise and aliasing, from a large sequence (e.g. hundreds or thousands) of RAW images taken with a consumer camera. This is a challenging issue requiring to perform on the fly demosaicking, denoising and super-resolution. Existing algorithms produce high-quality images but the number of input images is limited by severe computational and memory costs. In this thesis we propose an image fusion algorithm that processes the images sequentially so that the memory cost only depends on the size of the output image. After a preprocessing step, the mosaicked (or CFA) images are aligned in a common system of coordinates using a two-step registration method that we introduce. Then, a color image is computed by accumulation of the irregularly sampled data using classical kernel regression. Finally, the blur introduced is removed by applying the inverse of the corresponding asymptotic equivalent filter (that we introduce).We evaluate the performance and the accuracy of each step of our algorithm on synthetic and real data. We find that for a large sequence of RAW images, our method successfully performs super-resolution and the residual noise decreases as expected. We obtained results similar to those obtained by slower and memory greedy methods. As generating synthetic data requires an interpolation method, we also study in detail the trigonometric polynomial and B-spline interpolation methods. We derive from this study new fine-tuned interpolation methods
Chaabouni-Chouayakh, Houda. "Interprétation multi-niveaux des images RSO à haute résolution : application à l'analyse des zones urbaines à l'aide de techniques de fusion." Paris, ENST, 2009. http://www.theses.fr/2009ENST0024.
Повний текст джерелаWith the launch of the terrasar-x system, a new generation of high-resolution sar data is available. This opens new perspectives and challenges for the automatic interpretation of urban environments. A rich information content, previously hidden or not clearly distinguishable in low resolution images such as urban structures (small buildings, vehicles, etc), is now disclosed. However, only proper approaches are able to retrieve automatically this new detailed information. In fact, inside urban areas, the electromagnetic scattering is characterized by a variety of single or multiple scattering mechanisms with a wide range of scattering amplitudes. This makes the interpretation and information extraction over such areas from sar images quite complex to perform. This thesis provides solutions for the semi-automatic interpretation and mapping of urban areas using the high resolution provided by terrasar-x data. Our solutions take into account the raise of new man-made structures whose visibility has increased with the high resolution, and scattering response has improved with the high frequency x-band sar sensor carried by the terrasar-x system. They are mainly based on two steps. Firstly, we adopt a multi-layer sar image interpretation approach, which extracts and describes three kinds of information : backscattering, statistical and geometrical information. Secondly, information fusion techniques are applied to optimally exploit the different layers in order to improve the mapping of urban areas
Rougier, Simon. "Apport des images satellites à très haute résolution spatiale couplées à des données géographiques multi-sources pour l’analyse des espaces urbains." Thesis, Strasbourg, 2016. http://www.theses.fr/2016STRAH019/document.
Повний текст джерелаClimate change presents cities with significant environmental challenges. Urban planners need decision-making tools and a better knowledge of their territory. One objective is to better understand the link between the grey and the green infrastructures in order to analyse and represent them. The second objective is to propose a methodology to map the urban structure at urban fabric scale taking into account the grey and green infrastructures. In current databases, vegetation is not mapped in an exhaustive way. Therefore the first step is to extract tree and grass vegetation using Pléiades satellite images using an object-based image analysis and an active learning classification. Based on those classifications and multi-sources data, an approach based on knowledge discovery in databases is proposed. It is focused on set of indicators mostly coming from urbanism and landscape ecology. The methodology is built on Strasbourg and applied on Rennes to validate and check its reproducibility
Rougier, Simon. "Apport des images satellites à très haute résolution spatiale couplées à des données géographiques multi-sources pour l’analyse des espaces urbains." Electronic Thesis or Diss., Strasbourg, 2016. http://www.theses.fr/2016STRAH019.
Повний текст джерелаClimate change presents cities with significant environmental challenges. Urban planners need decision-making tools and a better knowledge of their territory. One objective is to better understand the link between the grey and the green infrastructures in order to analyse and represent them. The second objective is to propose a methodology to map the urban structure at urban fabric scale taking into account the grey and green infrastructures. In current databases, vegetation is not mapped in an exhaustive way. Therefore the first step is to extract tree and grass vegetation using Pléiades satellite images using an object-based image analysis and an active learning classification. Based on those classifications and multi-sources data, an approach based on knowledge discovery in databases is proposed. It is focused on set of indicators mostly coming from urbanism and landscape ecology. The methodology is built on Strasbourg and applied on Rennes to validate and check its reproducibility
Lopez-Ornelas, Erick de Jesus. "Segmentation d'images satellitaires à haute résolution et interaction avec l'information géographique : application à l'extraction de connaissances." Toulouse 3, 2005. http://www.theses.fr/2005TOU30021.
Повний текст джерелаTeina, Raimana. "Caractérisation de la cocoterai des Tuamotu à partir d'images satellites à très haute résolution spatiale." Paris 6, 2009. https://tel.archives-ouvertes.fr/tel-00516952.
Повний текст джерелаFerreira, Júlio César. "Algorithms for super-resolution of images based on sparse representation and manifolds." Thesis, Rennes 1, 2016. http://www.theses.fr/2016REN1S031/document.
Повний текст джерелаImage super-resolution is defined as a class of techniques that enhance the spatial resolution of images. Super-resolution methods can be subdivided in single and multi image methods. This thesis focuses on developing algorithms based on mathematical theories for single image super-resolution problems. Indeed, in order to estimate an output image, we adopt a mixed approach: i.e., we use both a dictionary of patches with sparsity constraints (typical of learning-based methods) and regularization terms (typical of reconstruction-based methods). Although the existing methods already perform well, they do not take into account the geometry of the data to: regularize the solution, cluster data samples (samples are often clustered using algorithms with the Euclidean distance as a dissimilarity metric), learn dictionaries (they are often learned using PCA or K-SVD). Thus, state-of-the-art methods still suffer from shortcomings. In this work, we proposed three new methods to overcome these deficiencies. First, we developed SE-ASDS (a structure tensor based regularization term) in order to improve the sharpness of edges. SE-ASDS achieves much better results than many state-of-the-art algorithms. Then, we proposed AGNN and GOC algorithms for determining a local subset of training samples from which a good local model can be computed for reconstructing a given input test sample, where we take into account the underlying geometry of the data. AGNN and GOC methods outperform spectral clustering, soft clustering, and geodesic distance based subset selection in most settings. Next, we proposed aSOB strategy which takes into account the geometry of the data and the dictionary size. The aSOB strategy outperforms both PCA and PGA methods. Finally, we combine all our methods in a unique algorithm, named G2SR. Our proposed G2SR algorithm shows better visual and quantitative results when compared to the results of state-of-the-art methods
Malaplate, Alain. "Radiométrie infrarouge : Développement et validation de méthodes utilisant la bande [3-5um] pour la détermination des paramètres de surface à haute résolution spatiale." Université Louis Pasteur (Strasbourg) (1971-2008), 2001. http://www.theses.fr/2001STR13227.
Повний текст джерелаBevilacqua, Marco. "Algorithms for super-resolution of images and videos based on learning methods." Phd thesis, Université Rennes 1, 2014. http://tel.archives-ouvertes.fr/tel-01064396.
Повний текст джерелаLegai, Pascal. "Impact de l'imagerie spatiale commerciale à haute résolution sur la sécurité internationale dans sa dimension de défense : perspectives pour l'Europe de la défense." Université de Marne-la-Vallée, 2003. http://www.theses.fr/2003MARN0153.
Повний текст джерелаPham, Thi Thanh Hien. "Développement des indicateurs de la qualité de vie urbaine à l'aide de la télédétection à très haute résolution spatiale cas de la ville de Hanoi." Thèse, Université de Sherbrooke, 2010. http://savoirs.usherbrooke.ca/handle/11143/2817.
Повний текст джерелаTremblais, Benoit. "De la résolution numérique des EDP à l'extraction de caractéristiques linéiques dans les images : application à la détection multi-échelles d'un arbre vasculaire." Phd thesis, Université de Poitiers, 2002. http://tel.archives-ouvertes.fr/tel-00349464.
Повний текст джерелаTremblais, Benoît. "De la résolution numérique des EDP à l'extraction de caractéristiques linéiques dans les images : application à la détection multi-échelles d'un arbre vasculaire." Poitiers, 2002. http://www.theses.fr/2002POIT2316.
Повний текст джерела