Rozprawy doktorskie na temat „Représentations robustes”
Utwórz poprawne odniesienie w stylach APA, MLA, Chicago, Harvard i wielu innych
Sprawdź 23 najlepszych rozpraw doktorskich naukowych na temat „Représentations robustes”.
Przycisk „Dodaj do bibliografii” jest dostępny obok każdej pracy w bibliografii. Użyj go – a my automatycznie utworzymy odniesienie bibliograficzne do wybranej pracy w stylu cytowania, którego potrzebujesz: APA, MLA, Harvard, Chicago, Vancouver itp.
Możesz również pobrać pełny tekst publikacji naukowej w formacie „.pdf” i przeczytać adnotację do pracy online, jeśli odpowiednie parametry są dostępne w metadanych.
Przeglądaj rozprawy doktorskie z różnych dziedzin i twórz odpowiednie bibliografie.
Morchid, Mohamed. "Représentations robustes de documents bruités dans des espaces homogènes". Thesis, Avignon, 2014. http://www.theses.fr/2014AVIG0202/document.
Pełny tekst źródłaIn the Information Retrieval field, documents are usually considered as a "bagof-words". This model does not take into account the temporal structure of thedocument and is sensitive to noises which can alter its lexical form. These noisescan be produced by different sources : uncontrolled form of documents in microbloggingplatforms, automatic transcription of speech documents which are errorprone,lexical and grammatical variabilities in Web forums. . . The work presented inthis thesis addresses issues related to document representations from noisy sources.The thesis consists of three parts in which different representations of content areavailable. The first one compares a classical representation based on a term-frequencyrepresentation to a higher level representation based on a topic space. The abstractionof the document content allows us to limit the alteration of the noisy document byrepresenting its content with a set of high-level features. Our experiments confirm thatmapping a noisy document into a topic space allows us to improve the results obtainedduring different information retrieval tasks compared to a classical approach based onterm frequency. The major problem with such a high-level representation is that it isbased on a space theme whose parameters are chosen empirically.The second part presents a novel representation based on multiple topic spaces thatallow us to solve three main problems : the closeness of the subjects discussed in thedocument, the tricky choice of the "right" values of the topic space parameters and therobustness of the topic-based representation. Based on the idea that a single representationof the contents cannot capture all the relevant information, we propose to increasethe number of views on a single document. This multiplication of views generates "artificial"observations that contain fragments of useful information. The first experimentvalidated the multi-view approach to represent noisy texts. However, it has the disadvantageof being very large and redundant and of containing additional variability associatedwith the diversity of views. In the second step, we propose a method based onfactor analysis to compact the different views and to obtain a new robust representationof low dimension which contains only the informative part of the document whilethe noisy variabilities are compensated. During a dialogue classification task, the compressionprocess confirmed that this compact representation allows us to improve therobustness of noisy document representation.Nonetheless, during the learning process of topic spaces, the document is consideredas a "bag-of-words" while many studies have showed that the word position in a7document is useful. A representation which takes into account the temporal structureof the document based on hyper-complex numbers is proposed in the third part. Thisrepresentation is based on the hyper-complex numbers of dimension four named quaternions.Our experiments on a classification task have showed the effectiveness of theproposed approach compared to a conventional "bag-of-words" representation
Paulin, Mattis. "De l'apprentissage de représentations visuelles robustes aux invariances pour la classification et la recherche d'images". Thesis, Université Grenoble Alpes (ComUE), 2017. http://www.theses.fr/2017GREAM007/document.
Pełny tekst źródłaThis dissertation focuses on designing image recognition systems which are robust to geometric variability. Image understanding is a difficult problem, as images are two-dimensional projections of 3D objects, and representations that must fall into the same category, for instance objects of the same class in classification can display significant differences. Our goal is to make systems robust to the right amount of deformations, this amount being automatically determined from data. Our contributions are twofolds. We show how to use virtual examples to enforce robustness in image classification systems and we propose a framework to learn robust low-level descriptors for image retrieval. We first focus on virtual examples, as transformation of real ones. One image generates a set of descriptors –one for each transformation– and we show that data augmentation, ie considering them all as iid samples, is the best performing method to use them, provided a voting stage with the transformed descriptors is conducted at test time. Because transformations have various levels of information, can be redundant, and can even be harmful to performance, we propose a new algorithm able to select a set of transformations, while maximizing classification accuracy. We show that a small amount of transformations is enough to considerably improve performance for this task. We also show how virtual examples can replace real ones for a reduced annotation cost. We report good performance on standard fine-grained classification datasets. In a second part, we aim at improving the local region descriptors used in image retrieval and in particular to propose an alternative to the popular SIFT descriptor. We propose new convolutional descriptors, called patch-CKN, which are learned without supervision. We introduce a linked patch- and image-retrieval dataset based on structure from motion of web-crawled images, and design a method to accurately test the performance of local descriptors at patch and image levels. Our approach outperforms both SIFT and all tested approaches with convolutional architectures on our patch and image benchmarks, as well as several styate-of-theart datasets
Barbano, Carlo Alberto Maria. "Collateral-Free Learning of Deep Representations : From Natural Images to Biomedical Applications". Electronic Thesis or Diss., Institut polytechnique de Paris, 2023. http://www.theses.fr/2023IPPAT038.
Pełny tekst źródłaDeep Learning (DL) has become one of the predominant tools for solving a variety of tasks, often with superior performance compared to previous state-of-the-art methods. DL models are often able to learn meaningful and abstract representations of the underlying data. However, it has been shown that they might also learn additional features, which are not necessarily relevant or required for the desired task. This could pose a number of issues, as this additional information can contain bias, noise, or sensitive information, that should not be taken into account (e.g. gender, race, age, etc.) by the model. We refer to this information as collateral. The presence of collateral information translates into practical issues when deploying DL-based pipelines, especially if they involve private users' data. Learning robust representations that are free of collateral information can be highly relevant for a variety of fields and applications, like medical applications and decision support systems.In this thesis, we introduce the concept of Collateral Learning, which refers to all those instances in which a model learns more information than intended. The aim of Collateral Learning is to bridge the gap between different fields in DL, such as robustness, debiasing, generalization in medical imaging, and privacy preservation. We propose different methods for achieving robust representations free of collateral information. Some of our contributions are based on regularization techniques, while others are represented by novel loss functions.In the first part of the thesis, we lay the foundations of our work, by developing techniques for robust representation learning on natural images. We focus on one of the most important instances of Collateral Learning, namely biased data. Specifically, we focus on Contrastive Learning (CL), and we propose a unified metric learning framework that allows us to both easily analyze existing loss functions, and derive novel ones. Here, we propose a novel supervised contrastive loss function, ε-SupInfoNCE, and two debiasing regularization techniques, EnD and FairKL, that achieve state-of-the-art performance on a number of standard vision classification and debiasing benchmarks.In the second part of the thesis, we focus on Collateral Learning in medical imaging, specifically on neuroimaging and chest X-ray images. For neuroimaging, we present a novel contrastive learning approach for brain age estimation. Our approach achieves state-of-the-art results on the OpenBHB dataset for age regression and shows increased robustness to the site effect. We also leverage this method to detect unhealthy brain aging patterns, showing promising results in the classification of brain conditions such as Mild Cognitive Impairment (MCI) and Alzheimer's Disease (AD). For chest X-ray images (CXR), we will target Covid-19 classification, showing how Collateral Learning can effectively hinder the reliability of such models. To tackle such issue, we propose a transfer learning approach that, combined with our regularization techniques, shows promising results on an original multi-site CXRs dataset.Finally, we provide some hints about Collateral Learning and privacy preservation in DL models. We show that some of our proposed methods can be effective in preventing certain information from being learned by the model, thus avoiding potential data leakage
Hafidi, Hakim. "Robust machine learning for Graphs/Networks". Electronic Thesis or Diss., Institut polytechnique de Paris, 2023. http://www.theses.fr/2023IPPAT004.
Pełny tekst źródłaThis thesis addresses advancements in graph representation learning, focusing on the challengesand opportunities presented by Graph Neural Networks (GNNs). It highlights the significanceof graphs in representing complex systems and the necessity of learning node embeddings that capture both node features and graph structure. The study identifies key issues in GNNs, such as their dependence on high-quality labeled data, inconsistent performanceacross various datasets, and susceptibility to adversarial attacks.To tackle these challenges, the thesis introduces several innovative approaches. Firstly, it employs contrastive learning for node representation, enabling self-supervised learning that reduces reliance on labeled data. Secondly, a Bayesian-based classifier isproposed for node classification, which considers the graph’s structure to enhance accuracy. Lastly, the thesis addresses the vulnerability of GNNs to adversarialattacks by assessing the robustness of the proposed classifier and introducing effective defense mechanisms.These contributions aim to improve both the performance and resilience of GNNs in graph representation learning
Roussillon, Tristan. "Algorithmes d'extraction de modèles géométriques discrets pour la représentation robuste des formes". Thesis, Lyon 2, 2009. http://www.theses.fr/2009LYO20103/document.
Pełny tekst źródłaThe work presented in this thesis concerns the fields of image analysis and discrete geometry. Image analysis aims at automatically describing the visual content of a digital image and discrete geometry provides tools devoted to digital image processing. A two-dimensional analog signal is regularly sampled in order to be handled on computers. This acquisition process results in a digital image, which is made up of a finite set of discrete elements. The topic of discrete geometry is to study the geometric properties of such kind of discrete spaces. In this work, we consider homogeneous regions of an image having a meaning for a user. The objective is to represent their digital contour by means of geometric patterns and compute measures. The scope of applications is wide in image analysis. For instance, our results would be of great interest for segmentation or object recognition. We focus on three discrete geometric patterns defined by Gauss digitization: the convex or concave part, the digital straight segment and the digital circular arc. We present several algorithms that detect or recognize these patterns on a digital contour. These algorithms are on-line, exact (integer-only computations without any approximation error) and fast (simplified computations thanks to arithmetic properties and linear-time complexity). They provide a way for segmenting a digital contour or for representing a digital contour by a reversible polygon. Moreover, we define a measure of convexity, a measure of straightness and a measure of circularity. These measures fulfil the following important properties: they are robust to rigid transformations, they may be applied on any part of a digital contour, they reach their maximal value for the template with which the data are compared to. From these measures, we introduce new patterns having a parameter that ranges from 0 to 1. The parameter is set to 1 when the localisation of the digital contour is reliable, but is set to a lower value when the digital contour is expected to have been shifted because of some acquisition noise. This measure-based approach provides a way for robustly decomposing a digital contour into convex, concave or straight parts
El, jili Fatimetou. "Représentation de signaux robuste aux bruits - Application à la détection et l'identification des signaux d'alarme". Thesis, Reims, 2018. http://www.theses.fr/2018REIMS040/document.
Pełny tekst źródłaThis work targets the detection and identification of audio signals and in particular alarm signals from priority cars. First, we propose a method for detecting alarm signals in a noisy environment, based on time-frequency signal analysis. This method makes it possible to detect and identify alarm signals embedded in noise, even with negative signal-to-noise ratios. Then we propose a signal quantization robust against transmission noise. This involves replacing each bit level of a vector of time or frequency samples with a binary word of the same length provided by an error- correcting encoder. In a first approach, each bit level is quantized independently of the others according to the Hamming distance minimization criterion. In a second approach, to reduce the quantization error at equal robustness, the different bit levels are quantized successively by a matching pursuit algorithm. This quantization gives the signals a specific shape that allows them to be easily recognized among other signals. Finally, we propose two methods for detecting and identifying signals based on robust quantization, operating in the time domain or in the frequency domain, by minimizing the distance between the received signals restricted to their high-weight bits and the reference signals. These methods make it possible to detect and identify signals in environments with very low signal-to-noise ratios, thanks to quantization. In addition, the first method, based on the time-frequency signature, is more efficient with quantized signals
Boimond, Jean-Louis. "Commande à modèle interne en représentation d'état. : Problèmes de synthèse d'algorithme de commande". Lyon, INSA, 1990. http://www.theses.fr/1990ISAL0102.
Pełny tekst źródła[The works presented in this thesis concern the Internal Model Control (I. M. C. ). The first part presents the main properties of this structure which combines the advantages of open-loop scheme (the controller is an approximate inverse of the model) and closed-loop structure (ability to cope with modelling errors and unmeasured disturbances). A comparison with the conventional closed-loop is briefly presented. In the second part, an asymptotic precision criterion is introduced; The conditions that are to be verified by the blocks of the I. M. C. , for zeroing the asymptotic error between the output and a polynomial input, are settled down. The controller is interpreted as an approximate inverse of the model. In discrete time, the use of F. I. R. (Finite Impulse Response) forms permits the synthesis of a stable and realisable controller. The third part deals with the problem of the model inversion in discrete time and in state space. It allows us to consider some vary linear or non-linear models, which are linear versus the control variable. The controller is decomposed in two parts: the first one generates the control variable in terms of model state and the reference objective, the second one generates the prediction of the reference signal. Asymptotic accuracy is guaranteed for reference inputs that are polynomial, with a given order, versus time. The last part presents the synthesis of an I. M. C. Based on the use of the above controller. The robustness filter becomes a predictor of the error between plant and model outputs, the dynamic of which is tuned according to the knowledge of the plant-model mismatch. Two approaches have been proposed to built in this filter. The first one uses the same technique as for the reference predictor. In the other, the usual notion of filtering is replaced by a measure of the prediction quality. ]
Oneata, Dan. "Modèles robustes et efficaces pour la reconnaissance d'action et leur localisation". Thesis, Université Grenoble Alpes (ComUE), 2015. http://www.theses.fr/2015GREAM019/document.
Pełny tekst źródłaVideo interpretation and understanding is one of the long-term research goals in computer vision. Realistic videos such as movies present a variety of challenging machine learning problems, such as action classification/action retrieval, human tracking, human/object interaction classification, etc. Recently robust visual descriptors for video classification have been developed, and have shown that it is possible to learn visual classifiers in realistic difficult settings. However, in order to deploy visual recognition systems on large-scale in practice it becomes important to address the scalability of the techniques. The main goal is this thesis is to develop scalable methods for video content analysis (eg for ranking, or classification)
Tran, Thi Quynh Nhi. "Robust and comprehensive joint image-text representations". Thesis, Paris, CNAM, 2017. http://www.theses.fr/2017CNAM1096/document.
Pełny tekst źródłaThis thesis investigates the joint modeling of visual and textual content of multimedia documents to address cross-modal problems. Such tasks require the ability to match information across modalities. A common representation space, obtained by eg Kernel Canonical Correlation Analysis, on which images and text can be both represented and directly compared is a generally adopted solution.Nevertheless, such a joint space still suffers from several deficiencies that may hinder the performance of cross-modal tasks. An important contribution of this thesis is therefore to identify two major limitations of such a space. The first limitation concerns information that is poorly represented on the common space yet very significant for a retrieval task. The second limitation consists in a separation between modalities on the common space, which leads to coarse cross-modal matching. To deal with the first limitation concerning poorly-represented data, we put forward a model which first identifies such information and then finds ways to combine it with data that is relatively well-represented on the joint space. Evaluations on emph{text illustration} tasks show that by appropriately identifying and taking such information into account, the results of cross-modal retrieval can be strongly improved. The major work in this thesis aims to cope with the separation between modalities on the joint space to enhance the performance of cross-modal tasks.We propose two representation methods for bi-modal or uni-modal documents that aggregate information from both the visual and textual modalities projected on the joint space. Specifically, for uni-modal documents we suggest a completion process relying on an auxiliary dataset to find the corresponding information in the absent modality and then use such information to build a final bi-modal representation for a uni-modal document. Evaluations show that our approaches achieve state-of-the-art results on several standard and challenging datasets for cross-modal retrieval or bi-modal and cross-modal classification
Siméoni, Oriane. "Robust image representation for classification, retrieval and object discovery". Thesis, Rennes 1, 2020. https://ged.univ-rennes1.fr/nuxeo/site/esupversions/415eb65b-d5f7-4be7-85e6-c2ecb2aba4dc.
Pełny tekst źródłaNeural network representations proved to be relevant for many computer vision tasks such as image classification, object detection, segmentation or instance-level image retrieval. A network is trained for one particular task and requires a large number of labeled data. We propose in this thesis solutions to extract the most information with the least supervision. First focusing on the classification task, we examine the active learning process in the context of deep learning and show that combining it to semi-supervised and unsupervised techniques boost greatly results. We then investigate the image retrieval task, and in particular we exploit the spatial localization information available ``for free'' in CNN feature maps. We first propose to represent an image by a collection of affine local features detected within activation maps, which are memory-efficient and robust enough to perform spatial matching. Then again extracting information from feature maps, we discover objects of interest in images of a dataset and gather their representations in a nearest neighbor graph. Using the centrality measure on the graph, we are able to construct a saliency map per image which focuses on the repeating objects and allows us to compute a global representation excluding clutter and background
Tran, Thi Quynh Nhi. "Robust and comprehensive joint image-text representations". Electronic Thesis or Diss., Paris, CNAM, 2017. http://www.theses.fr/2017CNAM1096.
Pełny tekst źródłaThis thesis investigates the joint modeling of visual and textual content of multimedia documents to address cross-modal problems. Such tasks require the ability to match information across modalities. A common representation space, obtained by eg Kernel Canonical Correlation Analysis, on which images and text can be both represented and directly compared is a generally adopted solution.Nevertheless, such a joint space still suffers from several deficiencies that may hinder the performance of cross-modal tasks. An important contribution of this thesis is therefore to identify two major limitations of such a space. The first limitation concerns information that is poorly represented on the common space yet very significant for a retrieval task. The second limitation consists in a separation between modalities on the common space, which leads to coarse cross-modal matching. To deal with the first limitation concerning poorly-represented data, we put forward a model which first identifies such information and then finds ways to combine it with data that is relatively well-represented on the joint space. Evaluations on emph{text illustration} tasks show that by appropriately identifying and taking such information into account, the results of cross-modal retrieval can be strongly improved. The major work in this thesis aims to cope with the separation between modalities on the joint space to enhance the performance of cross-modal tasks.We propose two representation methods for bi-modal or uni-modal documents that aggregate information from both the visual and textual modalities projected on the joint space. Specifically, for uni-modal documents we suggest a completion process relying on an auxiliary dataset to find the corresponding information in the absent modality and then use such information to build a final bi-modal representation for a uni-modal document. Evaluations show that our approaches achieve state-of-the-art results on several standard and challenging datasets for cross-modal retrieval or bi-modal and cross-modal classification
Chenot, Cécile. "Parcimonie, diversité morphologique et séparation robuste de sources". Thesis, Université Paris-Saclay (ComUE), 2017. http://www.theses.fr/2017SACLS223/document.
Pełny tekst źródłaThis manuscript addresses the Blind Source Separation (BSS) problem in the presence of outliers. Most BSS techniques are hampered by the presence of structured deviations from the standard linear mixing model, such as unexpected physical events or malfunctions of sensors. We propose a new data model taking explicitly into account the deviations. The resulting joint estimation of the components is an ill-posed problem, tackled using sparse modeling. The latter is particularly efficient for solving robust BSS since it allows for a robust unmixing of the sources jointly with a precise separation of the components. These works are then extended for the estimation of spectral variability in the framework of terrestrial hyperspectral imaging. Numerical experiments highlight the robustness and reliability of the proposed algorithms in a wide range of settings, including the full-rank regime
Victor, Stéphane. "Identification par modèle non entier pour la poursuite robuste de trajectoire par platitude". Thesis, Bordeaux 1, 2010. http://www.theses.fr/2010BOR14115/document.
Pełny tekst źródłaThe general theme of the work enables to handle a system, from identification to robust control. Flatness principles tackle path planning unless knowing the system model, hence the system parameter identification necessity. The principal contribution of this thesis deal with system identification by non integer models and with robust path tracking by the use of flatness principles for fractional models.Chapter 1 recalls the definitions and properties of a fractional operator and also the various representation methods of a fractional system. The stability theorem is also brought to mind. Fractional polynomial and fractional polynomial matrice algebras are introduced for the extension of flatness principles for fractional systems.Chapter 2 is about non integer model identification. After a state of the art on system identification by non integer model. Two contexts are considered : in presence of white noise and of colored noise. In each situation, two optimal (in variance and bias sense) estimators are put forward : one, when considering a known model structure with fixed differentiating orders, and another one by combining nonlinear programming technics for the optimization of coefficients and differentiating orders.Chapter 3 establishes the extension of flatness principles to fractional systems. Flatness of linear fractional systems are studied while considering different approaches such as transfer functions or pseudo-state-space representations with polynomial matrices. Path tracking robustness is ensured with CRONE control. Simulation examples display theoretical developments on flatness through thermal diffusion on a metallic rod. Finally, Chapter 4 is devoted to validate the contributions to system identification, to trajectory planning and to robust path tracking on a real fractional system : a metallic rod submitted to a heat flux
Djeddi, Mounir. "APPROCHES POUR L'ANALYSE DES SIGNAUX A PHASE POLYNOMIALE DANS UN ENVIRONNEMENT NON GAUSSIEN". Phd thesis, Université Paris Sud - Paris XI, 2005. http://tel.archives-ouvertes.fr/tel-00010277.
Pełny tekst źródłaOk, David. "Mise en Correspondance Robuste et Détection d'Éléments Visuels Appliquées à l'Analyse de Façades". Phd thesis, Université Paris-Est, 2013. http://tel.archives-ouvertes.fr/tel-00844049.
Pełny tekst źródłaArezki, Yassir. "Algorithmes de références 'robustes' pour la métrologie dimensionnelle des surfaces asphériques et des surfaces complexes en optique". Thesis, Université Paris-Saclay (ComUE), 2019. http://www.theses.fr/2019SACLN058.
Pełny tekst źródłaAspheres and freeform surfaces are a very challenging class of optical elements. Their application has grown considerably in the last few years in imaging systems, astronomy, lithography, etc. The metrology for aspheres is very challenging, because of the high dynamic range of the acquired information and the traceability to the SI unit meter. Metrology should make use of the infinite norm; (Minimum Zone Method or Min-Max method) to calculate the envelope enclosing the points in the dataset by minimizing the difference between the maximum deviation and the minimum deviation between the surface and the dataset. This method grows in complexity as the number of points in the dataset increases, and the involved algorithms are non-deterministic. Despite the fact that this method works for simple geometries (lines, planes, circles, cylinders, cones and spheres) it is still a major challenge when used on complex geometries (asphere and freeform surfaces). Therefore, the main objective is to address this key challenge about the development of Min-Max fitting algorithms for both aspherical and freeform surfaces as well as least squares fitting algorithms, in order to provide robust reference algorithms for the large community involved in this domain. The reference algorithms to be developed should be evaluated and validated on several reference data (softgauges) that will be generated using reference data generators
Huang, Di. "Robust face recognition based on three dimensional data". Phd thesis, Ecole Centrale de Lyon, 2011. http://tel.archives-ouvertes.fr/tel-00693158.
Pełny tekst źródłaDamerval, Christophe. "Ondelettes pour la détection de caractéristiques en traitement d'images. Application à la détection de région d'intérêt". Phd thesis, Grenoble 1, 2008. http://www.theses.fr/2008GRE10050.
Pełny tekst źródłaThis thesis in image processing addresses the problem of the highlight of some remarquable structures, such as objects we perceive visually. These can be monodimensional, like contours, as well as bidimensional, corresponding to more complex objects. An important problem in computer vision consists on detecting such structures, and also extracting characteristic features from them. In many applications, such as object recognition, image matching, motion tracking or the enhancement of some particular elements, it is a first step before other high-level operations. Thereby, the formulation of performant detectors appears as essential. We show that this can be carried out using wavelet decompositions; in particular, it is possible to define some maxima lines, which turn out as relevant to this problem : one the one hand, so as to detect objects (given by some regions of interest), and, on the other hand, in order to characterize them (computations of Lipschitz regularity and of characteristic scale). This original approach for detection, based on maxima lines, can thus be compared to classical approches
Damerval, Christophe. "Ondelettes pour la détection de caractéristiques en traitement d'images. Application à la détection de région d'intérêt". Phd thesis, Université Joseph Fourier (Grenoble), 2008. http://tel.archives-ouvertes.fr/tel-00287008.
Pełny tekst źródłaBernhardt, Stéphanie. "Performances et méthodes pour l'échantillonnage comprimé : Robustesse à la méconnaissance du dictionnaire et optimisation du noyau d'échantillonnage". Thesis, Université Paris-Saclay (ComUE), 2016. http://www.theses.fr/2016SACLS443/document.
Pełny tekst źródłaIn this thesis, we are interested in two different low rate sampling schemes that challenge Shannon’s theory: the sampling of finite rate of innovation signals and compressed sensing.Recently it has been shown that using appropriate sampling kernel, finite rate of innovation signals can be perfectly sampled even though they are non-bandlimited. In the presence of noise, reconstruction is achieved by a model-based estimation procedure. In this thesis, we consider the estimation of the amplitudes and delays of a finite stream of Dirac pulses using an arbitrary kernel and the estimation of a finite stream of arbitrary pulses using the Sum of Sincs (SoS) kernel. In both scenarios, we derive the Bayesian Cramér-Rao Bound (BCRB) for the parameters of interest. The SoS kernel is an interesting kernel since it is totally configurable by a vector of weights. In the first scenario, based on convex optimization tools, we propose a new kernel minimizing the BCRB on the delays, while in the second scenario we propose a family of kernels which maximizes the Bayesian Fisher Information, i.e., the total amount of information about each of the parameter in the measures. The advantage of the proposed family is that it can be user-adjusted to favor either of the estimated parameters.Compressed sensing is a promising emerging domain which outperforms the classical limit of the Shannon sampling theory if the measurement vector can be approximated as the linear combination of few basis vectors extracted from a redundant dictionary matrix. Unfortunately, in realistic scenario, the knowledge of this basis or equivalently of the entire dictionary is often uncertain, i.e. corrupted by a Basis Mismatch (BM) error. The related estimation problem is based on the matching of continuous parameters of interest to a discretized parameter set over a regular grid. Generally, the parameters of interest do not lie in this grid and there exists an estimation error even at high Signal to Noise Ratio (SNR). This is the off-grid (OG) problem. The consequence of the BM and the OG mismatch problems is that the estimation accuracy in terms of Bayesian Mean Square Error (BMSE) of popular sparse-based estimators collapses even if the support is perfectly estimated and in the high Signal to Noise Ratio (SNR) regime. This saturation effect considerably limits the effective viability of these estimation schemes.In this thesis, the BCRB is derived for CS model with unstructured BM and OG. We show that even though both problems share a very close formalism, they lead to different performances. In the biased dictionary based estimation context, we propose and study analytical expressions of the Bayesian Mean Square Error (BMSE) on the estimation of the grid error at high SNR. We also show that this class of estimators is efficient and thus reaches the Bayesian Cramér-Rao Bound (BCRB) at high SNR. The proposed results are illustrated in the context of line spectra analysis for several popular sparse estimator. We also study the Expected Cramér-Rao Bound (ECRB) on the estimation of the amplitude for a small OG error and show that it follows well the behavior of practical estimators in a wide SNR range.In the context of BM and OG errors, we propose two new estimation schemes called Bias-Correction Estimator (BiCE) and Off-Grid Error Correction (OGEC) respectively and study their statistical properties in terms of theoretical bias and variances. Both estimators are essentially based on an oblique projection of the measurement vector and act as a post-processing estimation layer for any sparse-based estimator and mitigate considerably the BM (OG respectively) degradation. The proposed estimators are generic since they can be associated to any sparse-based estimator, fast, and have good statistical properties. To illustrate our results and propositions, they are applied in the challenging context of the compressive sampling of finite rate of innovation signals
Belmerhnia, Leïla. "Approches parcimonieuses pour la sélection de variables et la classification : application à la spectroscopie IR de déchets de bois". Thesis, Université de Lorraine, 2017. http://www.theses.fr/2017LORR0039/document.
Pełny tekst źródłaIn this thesis, innovative techniques for sorting wood wastes are developed. The idea is to combine infrared spectrometry techniques with robust data processing methods for classification task. After exposing the context of the work in the first chapter, a state of the art on the spectral data classification is presented in the chapter 2. The third chapter deals with variable selection problem using sparse approaches. In particular we propose to extend some greedy methods for the simultaneous sparse approximation. The simulations performed for the approximation of an observation matrix validate the advantages of the proposed approaches. In the fourth chapter, we develop variable selection methods based on simultaneous sparse and regularized representation, to increase the performances of SVM classifier for the classification of NIR spectra and hyperspectral images of wood wastes. In the final chapter, we present the improvements made to the existing sorting systems. The results of the conducted tests using the processing software confirm that significant benefits can be achieved in terms of recycled wood quantities
Vaiter, Samuel. "Régularisations de Faible Complexité pour les Problèmes Inverses". Phd thesis, Université Paris Dauphine - Paris IX, 2014. http://tel.archives-ouvertes.fr/tel-01026398.
Pełny tekst źródła"Identification par modèle non entier pour la poursuite robuste de trajectoire par platitude". 2010. http://tel.archives-ouvertes.fr/tel-00978545.
Pełny tekst źródła