Descripteur
Documents disponibles dans cette catégorie (51)
Ajouter le résultat dans votre panier Affiner la recherche Interroger des sources externes
Etendre la recherche sur niveau(x) vers le bas
Multiscale supervised kernel dictionary learning for SAR target recognition / Lei Tao in IEEE Transactions on geoscience and remote sensing, vol 58 n° 9 (September 2020)
[article]
Titre : Multiscale supervised kernel dictionary learning for SAR target recognition Type de document : Article/Communication Auteurs : Lei Tao, Auteur ; Xue Jiang, Auteur ; Xingzhao Liu, Auteur ; et al., Auteur Année de publication : 2020 Article en page(s) : pp 6281 - 6297 Note générale : bibliographie Langues : Anglais (eng) Descripteur : [Vedettes matières IGN] Traitement d'image radar et applications
[Termes IGN] analyse en composantes principales
[Termes IGN] apprentissage automatique
[Termes IGN] classification dirigée
[Termes IGN] classification par réseau neuronal convolutif
[Termes IGN] détection automatique
[Termes IGN] détection de cible
[Termes IGN] erreur de classification
[Termes IGN] image radar moirée
[Termes IGN] méthode fondée sur le noyau
[Termes IGN] reconstruction d'imageRésumé : (auteur) In this article, a supervised nonlinear dictionary learning (DL) method, called multiscale supervised kernel DL (MSK-DL), is proposed for target recognition in synthetic aperture radar (SAR) images. We use Frost filters with different parameters to extract an SAR image’s multiscale features for data augmentation and noise suppression. In order to reduce the computation cost, the dimension of each scale feature is reduced by principal component analysis (PCA). Instead of the widely used linear DL, we learn multiple nonlinear dictionaries to capture the nonlinear structure of data by introducing the dimension-reduced features into the nonlinear reconstruction error terms. A classification model, which is defined as a discriminative classification error term, is learned simultaneously. Hence, the objective function contains the nonlinear reconstruction error terms and a classification error term. Two optimization algorithms, called multiscale supervised kernel K-singular value decomposition (MSK-KSVD) and multiscale supervised incremental kernel DL (MSIK-DL), are proposed to compute the multidictionary and the classifier. Experiments on the moving and stationary target automatic recognition (MSTAR) data set are performed to evaluate the effectiveness of the two proposed algorithms. And the experimental results demonstrate that the proposed scheme outperforms some representative common machine learning strategies, state-of-the-art convolutional neural network (CNN) models and some representative DL methods, especially in terms of its robustness against training set size and noise. Numéro de notice : A2020-529 Affiliation des auteurs : non IGN Thématique : IMAGERIE/INFORMATIQUE Nature : Article nature-HAL : ArtAvecCL-RevueIntern DOI : 10.1109/TGRS.2020.2976203 Date de publication en ligne : 03/03/2020 En ligne : https://doi.org/10.1109/TGRS.2020.2976203 Format de la ressource électronique : URL article Permalink : https://documentation.ensg.eu/index.php?lvl=notice_display&id=95709
in IEEE Transactions on geoscience and remote sensing > vol 58 n° 9 (September 2020) . - pp 6281 - 6297[article]
Titre : Approche bayésienne pour la sélection de modèles : Application à la restauration d’image Type de document : Thèse/HDR Auteurs : Benjamin Harroué, Auteur ; Jean-François Giovannelli, Directeur de thèse ; Marcela Pereyra, Directeur de thèse Editeur : Bordeaux : Université de Bordeaux Année de publication : 2020 Importance : 102 p. Format : 21 x 30 cm Note générale : bibliographie
Thèse présentée pour obtenir le grade de Docteur en Automatique, Productique, Signal et Image, Ingénierie cognitiqueLangues : Français (fre) Descripteur : [Vedettes matières IGN] Traitement d'image optique
[Termes IGN] déconvolution
[Termes IGN] échantillonnage de Gibbs
[Termes IGN] estimation bayesienne
[Termes IGN] fonction harmonique
[Termes IGN] matrice de covariance
[Termes IGN] problème inverse
[Termes IGN] processus gaussien
[Termes IGN] reconstruction d'image
[Termes IGN] restauration d'imageIndex. décimale : THESE Thèses et HDR Résumé : (auteur) L’inversion consiste à reconstruire des objets d’intérêt à partir de données acquises au travers d’un système d’observation. Dans ces travaux, nous nous penchons sur la déconvolution d’image. Les données observées constituent une version dégradée de l’objet, altéré par le système (flou et bruit). A cause de la perte d’informations engendrée, le problème devient alors mal conditionné. Une solution est de régulariser dans un cadre bayésien : en se basant sur des modèles, on introduit de l’information a priori sur les inconnues. Se posent alors les questions suivantes : comment comparer les modèles candidats et choisir le meilleur ? Sur quel critère faut-il s’appuyer ? A quelles caractéristiques ou quantités doit-on se fier ? Ces travaux présentent une méthode de comparaison et de sélection automatique de modèles, fondée sur la théorie de la décision bayésienne. La démarche consiste à sélectionner le modèle qui maximise la probabilité a posteriori. Pour calculer ces dernières, on a besoin de connaître une quantité primordiale : l’évidence. Elle s’obtient en marginalisant la loi jointe par rapport aux inconnus : l’image et les hyperparamètres. Les dépendances complexes entre les variables et la grande dimension de l’image rendent le calcul analytique de l’intégrale impossible. On a donc recours à des méthodes numériques. Dans cette première étude, on s’intéresse au cas gaussien circulant. Cela permet, d’une part, d’avoir une expression analytique de l’intégrale sur l’image, et d’autre part, de faciliter la manipulation des matrices de covariances. Plusieurs méthodes sont mises en œuvre comme l’algorithme du Chib couplé à une chaîne de Gibbs, les power posteriors, ou encore la moyenne harmonique. Les méthodes sont ensuite comparées pour déterminer lesquelles sont les plus adéquates au problème de la restauration d’image. Note de contenu : 1- Introduction
2- Sélection de modèles et calcul de l’évidence : état de l’art
3- Sélection de modèles sur observation directe
4- Sélection de modèles sur observation indirecte
5- Sélection de modèles sur données réelles
6- Conclusion : bilan et perspectivesNuméro de notice : 28558 Affiliation des auteurs : non IGN Thématique : IMAGERIE Nature : Thèse française Note de thèse : thèse de Doctorat : Automatique, Productique, Signal et Image, Ingénierie cognitique : Bordeaux : 2020 nature-HAL : Thèse En ligne : https://tel.archives-ouvertes.fr/tel-03065948/document Format de la ressource électronique : URL Permalink : https://documentation.ensg.eu/index.php?lvl=notice_display&id=97587
Titre : Intelligent Imaging and Analysis Type de document : Monographie Auteurs : DaeEun Kim, Éditeur scientifique ; Dosik Hwang, Éditeur scientifique Editeur : Bâle [Suisse] : Multidisciplinary Digital Publishing Institute MDPI Année de publication : 2020 Importance : 492 p. Format : 16 x 24 cm ISBN/ISSN/EAN : 978-3-03921-921-6 Note générale : bibliographie Langues : Anglais (eng) Descripteur : [Vedettes matières IGN] Traitement d'image
[Termes IGN] apprentissage profond
[Termes IGN] classification par réseau neuronal convolutif
[Termes IGN] estimation de pose
[Termes IGN] image 3D
[Termes IGN] image captée par drone
[Termes IGN] imagerie médicale
[Termes IGN] reconstruction d'image
[Termes IGN] segmentation d'image
[Termes IGN] texture d'image
[Termes IGN] vision par ordinateurRésumé : (éditeur) Imaging and analysis are widely involved in various research fields, including biomedical applications, medical imaging and diagnosis, computer vision, autonomous driving, and robot controls. Imaging and analysis are now facing big changes regarding intelligence, due to the breakthroughs of artificial intelligence techniques, including deep learning. Many difficulties in image generation, reconstruction, de-noising skills, artifact removal, segmentation, detection, and control tasks are being overcome with the help of advanced artificial intelligence approaches. This Special Issue focuses on the latest developments of learning-based intelligent imaging techniques and subsequent analyses, which include photographic imaging, medical imaging, detection, segmentation, medical diagnosis, computer vision, and vision-based robot control. These latest technological developments will be shared through this Special Issue for the various researchers who are involved with imaging itself, or are using image data and analysis for their own specific purposes. Note de contenu : 1- Special features on intelligent imaging and analysis
2- Intelligent evaluation of strabismus in videos based on an automated cover test
3- Application of a real-time visualization method of AUVs in underwater visual localization
4- Volumetric tooth wear measurement of scraper conveyor sprocket using shape from
focus-based method
5- A novel self-intersection penalty term for statistical body shape models and its applications in 3D pose estimation
6- A CNN model for human parsing based on capacity optimization
7- Fast 3D semantic mapping in road scenes †
8- Automated classification analysis of geological structures based on images data and deep learning model
9- Dark spot detection in SAR images of oil spill using segnet
10- A high-resolution texture mapping technique for 3D textured model
11- Image super-resolution algorithm based on dual-channel convolutional neural networks
12- No-reference automatic quality assessment for colorfulness-adjusted, contrast-adjusted, and sharpness-adjusted images using high-dynamic-range-derived features
13- A novel one-camera-five-mirror three-dimensional imaging method for reconstructing the cavitation bubble cluster in a water hydraulic valve
14- Deep residual network with sparse feedback for image restoration
15- An image segmentation method using an active contour model based on improved SPF
and LIF
16- Image segmentation approaches for weld pool monitoring during robotic arc welding
17- A novel discriminating and relative global spatial image representation with applications in CBIR
18- Double low-rank and sparse decomposition for surface defect segmentation of steel sheet
19- A UAV-based visual inspection method for rail surface defects
20- Feature-learning-based printed circuit board inspection via speeded-up robust features and random forest
21- Research progress of visual inspection technology of steel products
22- Fine-grain segmentation of the intervertebral discs from MR spine images using deep convolutional neural networks: BSU-Net
23- Semi-automatic segmentation of vertebral bodies in MR images of human lumbar spinesNuméro de notice : 28500 Affiliation des auteurs : non IGN Thématique : IMAGERIE Nature : Monographie DOI : 10.3390/books978-3-03921-921-6 En ligne : https://doi.org/10.3390/books978-3-03921-921-6 Format de la ressource électronique : URL Permalink : https://documentation.ensg.eu/index.php?lvl=notice_display&id=96897
Titre : Learning 3D generation and matching Type de document : Thèse/HDR Auteurs : Thibault Groueix, Auteur ; Mathieu Aubry, Directeur de thèse Editeur : Paris : Ecole Nationale des Ponts et Chaussées ENPC Année de publication : 2020 Importance : 169 p. Format : 21 x 30 cm Note générale : bibliographie
A doctoral thesis in the domain of automated signal and image processing submitted to École Doctorale Paris-Est
Mathématiques et Sciences et Technologies de l’Information et de la CommunicationLangues : Anglais (eng) Descripteur : [Vedettes matières IGN] Traitement d'image optique
[Termes IGN] appariement de formes
[Termes IGN] appariement dense
[Termes IGN] apprentissage profond
[Termes IGN] classification par réseau neuronal convolutif
[Termes IGN] déformation de surface
[Termes IGN] isométrie
[Termes IGN] maillage
[Termes IGN] modélisation 3D
[Termes IGN] reconstruction 3D
[Termes IGN] reconstruction d'image
[Termes IGN] segmentation d'image
[Termes IGN] semis de points
[Termes IGN] voxelIndex. décimale : THESE Thèses et HDR Résumé : (auteur) The goal of this thesis is to develop deep learning approaches to model and analyse 3D shapes. Progress in this field could democratize artistic creation of 3D assets which currently requires time and expert skills with technical software. We focus on the design of deep learning solutions for two particular tasks, key to many 3D modeling applications: single-view reconstruction and shape matching. A single-view reconstruction (SVR) method takes as input a single image and predicts the physical world which produced that image. SVR dates back to the early days of computer vision. In particular, in the 1960s, Lawrence G. Roberts proposed to align simple 3D primitives to the input image under the assumption that the physical world is made of cuboids. Another approach proposed by Berthold Horn in the 1970s is to decompose the input image in intrinsic images and use those to predict the depth of every input pixel. Since several configurations of shapes, texture and illumination can explain the same image, both approaches need to form assumptions on the distribution of images and 3D shapes to resolve the ambiguity. In this thesis, we learn these assumptions from large-scale datasets instead of manually designing them. Learning allows us to perform complete object reconstruction, including parts which are not visible in the input image. Shape matching aims at finding correspondences between 3D objects. Solving this task requires both a local and global understanding of 3D shapes which is hard to achieve explicitly. Instead we train neural networks on large-scale datasets to solve this task and capture this knowledge implicitly through their internal parameters.Shape matching supports many 3D modeling applications such as attribute transfer, automatic rigging for animation, or mesh editing.The first technical contribution of this thesis is a new parametric representation of 3D surfaces modeled by neural networks.The choice of data representation is a critical aspect of any 3D reconstruction algorithm. Until recently, most of the approaches in deep 3D model generation were predicting volumetric voxel grids or point clouds, which are discrete representations. Instead, we present an alternative approach that predicts a parametric surface deformation ie a mapping from a template to a target geometry. To demonstrate the benefits of such a representation, we train a deep encoder-decoder for single-view reconstruction using our new representation. Our approach, dubbed AtlasNet, is the first deep single-view reconstruction approach able to reconstruct meshes from images without relying on an independent post-processing, and can do it at arbitrary resolution without memory issues. A more detailed analysis of AtlasNet reveals it also generalizes better to categories it has not been trained on than other deep 3D generation approaches.Our second main contribution is a novel shape matching approach purely based on reconstruction via deformations. We show that the quality of the shape reconstructions is critical to obtain good correspondences, and therefore introduce a test-time optimization scheme to refine the learned deformations. For humans and other deformable shape categories deviating by a near-isometry, our approach can leverage a shape template and isometric regularization of the surface deformations. As category exhibiting non-isometric variations, such as chairs, do not have a clear template, we learn how to deform any shape into any other and leverage cycle-consistency constraints to learn meaningful correspondences. Our reconstruction-for-matching strategy operates directly on point clouds, is robust to many types of perturbations, and outperforms the state of the art by 15% on dense matching of real human scans. Note de contenu : 1- Introduction
2 Related Work
3 AtlasNet: A Papier-Mache Approach to Learning 3D Surface Generation
4 3D-CODED : 3D Correspondences by Deep Deformation
5 Unsupervised cycle-consistent deformation for shape matching
6 ConclusionNuméro de notice : 28310 Affiliation des auteurs : non IGN Thématique : IMAGERIE Nature : Thèse française Note de thèse : Thèse de Doctorat : Automated signal and image processing : Paris-Est : 2020 Organisme de stage : LIGM DOI : sans En ligne : https://tel.archives-ouvertes.fr/tel-03127055v2/document Format de la ressource électronique : URL Permalink : https://documentation.ensg.eu/index.php?lvl=notice_display&id=98201
Titre : Learning stereo reconstruction with deep neural networks Type de document : Thèse/HDR Auteurs : Stepan Tulyakov, Auteur ; François Fleuret, Directeur de thèse ; Anton Ivanov, Directeur de thèse Editeur : Lausanne : Ecole Polytechnique Fédérale de Lausanne EPFL Année de publication : 2020 Importance : 139 p. Format : 21 x 30 cm Note générale : bibliographie
Thèse présentée à l'Ecole Polytechnique Fédérale de Lausanne pour l’obtention du grade de Docteur ès SciencesLangues : Anglais (eng) Descripteur : [Vedettes matières IGN] Traitement d'image optique
[Termes IGN] apprentissage profond
[Termes IGN] classification semi-dirigée
[Termes IGN] contrainte géométrique
[Termes IGN] couple stéréoscopique
[Termes IGN] entropie
[Termes IGN] estimateur
[Termes IGN] étalonnage géométrique
[Termes IGN] modèle stéréoscopique
[Termes IGN] profondeur
[Termes IGN] réalité de terrain
[Termes IGN] reconstruction 3D
[Termes IGN] reconstruction d'image
[Termes IGN] vision par ordinateur
[Termes IGN] vision stéréoscopiqueRésumé : (auteur) Stereo reconstruction is a problem of recovering a 3d structure of a scene from a pair of images of the scene, acquired from different viewpoints. It has been investigated for decades and many successful methods were developed. The main drawback of these methods, is that they typically utilize a single depth cue, such as parallax, defocus blur or shading, and thus are not as robust as a human visual system that simultaneously relies on a range of monocular and binocular cues. This is mainly because it is hard to manually design a model, accounting for multiple depth cues. In this work, we address this problem by focusing on deep learning-based stereo methods that can discover a model for multiple depth cues directly from training data with ground truth depth. The complexity of deep learning-based methods, however, requires very large training sets with ground truth depth, which is often hard or costly to collect. Furthermore, even when training data is available it is often contaminated with noise, which reduces the effectiveness of supervised learning. In this work, in Chapter 3 we show that it is possible to alleviate this problem by using weakly supervised learning, that utilizes geometric constraints of the problem instead of ground truth depth. Besides the large training set requirement, deep stereo methods are not as application-friendlyas traditional methods. They have a large memory footprint and their disparity range is fixed at training time. For some applications, such as satellite stereo i magery, these are serious problems since satellite images are very large, often reaching tens of megapixels, and have a variable baseline, depending on a time difference between stereo images acquisition. In this work, in Chapter 4 we address these problems by introducing a novel network architecture with a bottleneck, capable of processing large images and utilizing more context, and an estimator that makes the network less sensitive to stereo matching ambiguities and applicable to any disparity range without re-training. Because deep learning-based methods discover depth cues directly from training data, they can be adapted to new data modalities without large modifications. In this work, in Chapter 5 we show that our method, developed for a conventional frame-based camera, can be used with a novel event-based camera, that has a higher dynamic range, smaller latency, and low power consumption. Instead of sampling intensity of all pixels with a fixed frequency, this camera asynchronously reports events of significant pixel intensity changes. To adopt our method to this new data modality, we propose a novel event sequence embedding module, that firstly aggregates information locally, across time, using a novel fully-connected layer for an irregularly sampled continuous domain, and then across discrete spatial domain. One interesting application of stereo is a reconstruction of a planet’s surface topography from satellite stereo images. In this work, in Chapter 6 we describe a geometric calibration method, as well as mosaicing and stereo reconstruction tools that we developed in the framework of the doctoral project for Color and Stereo Surface Imaging System onboard of ESA’s Trace Gas Orbiter, orbiting Mars. For the calibration, we propose a novel method, relying on starfield images because large focal lengths and complex optical distortion of the instrument forbid using standard methods. Scientific and practical results of this work are widely used by a scientific community. Note de contenu : 1- Introduction
2- Background
3- Weakly supervised learning of deep patch-matching cost
4- Applications-friendly deep stereo
5- Dense deep event-based stereo
6- Calibration of a satellite stereo system
7- ConclusionsNuméro de notice : 25795 Affiliation des auteurs : non IGN Thématique : IMAGERIE Nature : Thèse étrangère Note de thèse : Thèse de Doctorat : Sciences : Lausanne : 2020 En ligne : https://infoscience.epfl.ch/record/275342?ln=fr Format de la ressource électronique : URL Permalink : https://documentation.ensg.eu/index.php?lvl=notice_display&id=95025 PermalinkInnovative techniques of photogrammetry for 3D modeling / Vicenzo Barrile in Applied geomatics, Vol 11 n° 4 (December 2019)PermalinkA CNN-based subpixel level DSM generation approach via single image super-resolution / Yongjun Zhang in Photogrammetric Engineering & Remote Sensing, PERS, vol 85 n° 10 (October 2019)PermalinkBayesian iterative reconstruction methods for 3D X-ray Computed Tomography / Camille Chapdelaine (2019)PermalinkPermalinkMarkov random field for combined defogging and stereo reconstruction / Laurent Caraffa (2018)PermalinkSparse signal modeling: Application to image compression, Image error concealment and compressed sensing / Ali Akbari (2018)PermalinkColour Helmholtz stereopsis for reconstruction of dynamic scenes with arbitrary unknown reflectance / Nadejda Roubtsova in International journal of computer vision, vol 124 n° 1 (August 2017)PermalinkCorrection du flou de mouvement sur des images prises de nuit depuis un véhicule de numérisation terrestre / Vincent Daval in Revue Française de Photogrammétrie et de Télédétection, n° 215 (mai - août 2017)PermalinkGeneralizing the prediction sum of squares statistic and formula, application to linear fractional image warp and surface fitting / Adrien Bartoli in International journal of computer vision, vol 122 n° 1 (March 2017)PermalinkNew point matching algorithm using sparse representation of image patch feature for SAR image registration / Jianwei Fan in IEEE Transactions on geoscience and remote sensing, vol 55 n° 3 (March 2017)PermalinkDictionary learning for promoting structured sparsity in hyperspectral compressive sensing / Lei Zhang in IEEE Transactions on geoscience and remote sensing, vol 54 n° 12 (December 2016)PermalinkObject-based morphological profiles for classification of remote sensing imagery / Christian Geiss in IEEE Transactions on geoscience and remote sensing, vol 54 n° 10 (October 2016)PermalinkAssessing the planimetric accuracy of historical maps (sixteenth to nineteenth centuries): new methods and potential for coastal landscape reconstruction / Iason Jongepier in Cartographic journal (the), Vol 53 n° 2 (May 2016)PermalinkExploiting joint sparsity for pansharpening : the J-SparseFI algorithm / Xiao Xiang Zhu in IEEE Transactions on geoscience and remote sensing, vol 54 n° 5 (May 2016)PermalinkBand-limited signal reconstruction from irregular samples with variable apertures / David G. Long in IEEE Transactions on geoscience and remote sensing, vol 54 n° 4 (April 2016)PermalinkThin cloud removal based on signal transmission principles and spectral mixture analysis / Meng Xu in IEEE Transactions on geoscience and remote sensing, vol 54 n° 3 (March 2016)PermalinkAirborne DLSLA 3-D SAR image reconstruction by combination of polar formatting and L_1 regularization / Xueming Peng in IEEE Transactions on geoscience and remote sensing, vol 54 n° 1 (January 2016)PermalinkHYCA: A new technique for hyperspectral compressive sensing / G. Martin in IEEE Transactions on geoscience and remote sensing, vol 53 n° 5 (mai 2015)PermalinkPermalink