Centre dedocumentation
scientifique

Bayesian transfer learning for object detection in optical remote sensing images / Changsheng Zhou in IEEE Transactions on geoscience and remote sensing, vol 58 n° 11 (November 2020)  

Public [article]  inIEEE Transactions on geoscience and remote sensing > vol 58 n° 11 (November 2020) . - pp 7705 - 7719 Titre : Bayesian transfer learning for object detection in optical remote sensing images Type de document : Article/Communication Auteurs : Changsheng Zhou, Auteur ; Jiangshe Zhang, Auteur ; Junmin Liu, Auteur ; et al., Auteur Année de publication : 2020 Article en page(s) : pp 7705 - 7719 Note générale : bibliographie Langues : Anglais (eng) Descripteur : [Vedettes matières IGN] Traitement d'image optique [Termes descripteurs IGN] apprentissage profond [Termes descripteurs IGN] chaîne de traitement [Termes descripteurs IGN] détection d'objet [Termes descripteurs IGN] distribution de Fisher [Termes descripteurs IGN] jeu de données localisées [Termes descripteurs IGN] théorème de Bayes Résumé : (auteur) In the literature of object detection in optical remote sensing images, a popular pipeline is first modifying an off-the-shelf deep neural network, then initializing the modified network by pretrained weights on a source data set, and finally fine-tuning the network on a target data set. The procedure works well in practice but might not make full use of underlying knowledge implied by pretrained weights. In this article, we propose a novel method, referred to as Fisher regularization, for efficient knowledge transferring. Based on Bayes’ theorem, the method stores underlying knowledge into a Fisher information matrix and fine-tunes parameters based on the knowledge. The proposed method would not introduce extra parameters and is less sensitive to hyperparameters than classical weight decay. Experiments on NWPUVHR-10 and DOTA data sets show that the proposed method is effective and works well with different object detectors. Numéro de notice : A2020-679 Affiliation des auteurs : non IGN Thématique : IMAGERIE Nature : Article nature-HAL : ArtAvecCL-RevueIntern DOI : 10.1109/TGRS.2020.2983201 date de publication en ligne : 14/04/2020 En ligne : https://doi.org/10.1109/TGRS.2020.2983201 Format de la ressource électronique : url article Permalink : https://documentation.ensg.eu/index.php?lvl=notice_display&id=96182 [article]

Bayesian iterative reconstruction methods for 3D X-ray Computed Tomography / Camille Chapdelaine (2019)  

Public Titre : Bayesian iterative reconstruction methods for 3D X-ray Computed Tomography Type de document : Thèse/HDR Auteurs : Camille Chapdelaine, Auteur ; Charles Soussen, Directeur de thèse Editeur : Paris-Orsay : Université de Paris 11 Paris-Sud Centre d'Orsay Année de publication : 2019 Importance : 185 p. Format : 21 x 30 cm Note générale : Bibliographie Thèse de Doctorat de l’Université Paris - Saclay préparée à l'Université Paris-Sud, Sciences et Technologies de l’Information et de la Communication (STIC), Traitement du signal et des images Langues : Anglais (eng) Descripteur : [Vedettes matières IGN] Traitement d'image [Termes descripteurs IGN] artefact [Termes descripteurs IGN] capteur-projecteur [Termes descripteurs IGN] faisceau [Termes descripteurs IGN] inférence [Termes descripteurs IGN] itération [Termes descripteurs IGN] processeur graphique [Termes descripteurs IGN] rayon X [Termes descripteurs IGN] reconstruction 3D [Termes descripteurs IGN] reconstruction d'image [Termes descripteurs IGN] segmentation [Termes descripteurs IGN] spectroscopie [Termes descripteurs IGN] théorème de Bayes [Termes descripteurs IGN] tomographie [Termes descripteurs IGN] volume (grandeur) Index. décimale : THESE Thèses et HDR Résumé : (auteur) In industry, 3D X-ray Computed Tomography aims at virtually imaging a volume in order to inspect its interior. The virtual volume is obtained thanks to a reconstruction algorithm based on projections of X-rays sent through the industrial part to inspect. In order to compensate uncertainties in the projections such as scattering or beam-hardening, which are cause of many artifacts in conventional filtered backprojection methods, iterative reconstruction methods bring further information by enforcing a prior model on the volume to reconstruct, and actually enhance the reconstruction quality. In this context, this thesis proposes new iterative reconstruction methods for the inspection of aeronautical parts made by SAFRAN group. In order to alleviate the computational cost due to repeated projection and backprojection operations which model the acquisition process, iterative reconstruction methods can take benefit from the use of high-parallel computing on Graphical Processor Unit (GPU). In this thesis, the implementation on GPU of several pairs of projector and backprojector is detailed. In particular, a new GPU implementation of the matched Separable Footprint pair is proposed. Since many of SAFRAN's industrial parts are piecewise-constant volumes, a Gauss-Markov-Potts prior model is introduced, from which a joint reconstruction and segmentation algorithm is derived. This algorithm is based on a Bayesian approach which enables to explain the role of each parameter. The actual polychromacy of X-rays, which is responsible for scattering and beam-hardening, is taken into account by proposing an error-splitting forward model. Combined with Gauss-Markov-Potts prior on the volume, this new forward model is experimentally shown to bring more accuracy and robustness. At last, the estimation of the uncertainties on the reconstruction is investigated by variational Bayesian approach. In order to have a reasonable computation time, it is highlighted that the use of a matched pair of projector and backprojector is necessary. Note de contenu : 1- X-ray computed tomography : an inverse problem 2- Reconstruction methods in X-ray computed tomography 3- Projection and backprojection operators 4- Gauss-Markov-Potts prior model for joint reconstruction and segmentation 5- Error-splitting forward model and its application with Gauss-Markov-Potts prior 6- Towards the estimation of the uncertainties on the reconstruction by Variational Bayesian Approach 7- Conclusion and perspectives Numéro de notice : 25702 Affiliation des auteurs : non IGN Thématique : IMAGERIE Nature : Thèse française Note de thèse : Thèse de Doctorat : Sciences et Technologies de l’Information et de la Communication (STIC) : Traitement du signal et des images : Paris 11 : 2019 Organisme de stage : Safran DOI : sans En ligne : https://tel.archives-ouvertes.fr/tel-02110033 Format de la ressource électronique : URL Permalink : https://documentation.ensg.eu/index.php?lvl=notice_display&id=94827

Improving the reliability of landslide susceptibility mapping through spatial uncertainty analysis: a case study of Al Hoceima, Northern Morocco / Hassane Rahali in Geocarto international, vol 34 n° 1 ([01/01/2019])  

Public [article]  inGeocarto international > vol 34 n° 1 [01/01/2019] . - pp 43 - 77 Titre : Improving the reliability of landslide susceptibility mapping through spatial uncertainty analysis: a case study of Al Hoceima, Northern Morocco Type de document : Article/Communication Auteurs : Hassane Rahali, Auteur Année de publication : 2019 Article en page(s) : pp 43 - 77 Note générale : bibliographie Langues : Anglais (eng) Descripteur : [Vedettes matières IGN] Analyse spatiale [Termes descripteurs IGN] analyse comparative [Termes descripteurs IGN] analyse des risques [Termes descripteurs IGN] effondrement de terrain [Termes descripteurs IGN] géomorphologie locale [Termes descripteurs IGN] incertitude géométrique [Termes descripteurs IGN] lithologie [Termes descripteurs IGN] Maroc [Termes descripteurs IGN] méthode de Monte-Carlo [Termes descripteurs IGN] méthode fiable [Termes descripteurs IGN] modèle de simulation [Termes descripteurs IGN] processus stochastique [Termes descripteurs IGN] régression logistique [Termes descripteurs IGN] théorème de Bayes [Termes descripteurs IGN] zone à risque Résumé : (auteur) This paper aims at providing an answer as to whether generalization obtained with data-driven modelling can be used to gauge the plausibility of the physically based (PB) model’s prediction. Two statistical models namely; Weight of Evidence (WofE) and Logistic Regression (LR), and a PB model using the infinite slope assumptions were evaluated and compared with respect to their abilities to predict susceptible areas to shallow landslides at the 1:10.000 urban scale. Threshold-dependent performance metrics showed that the three methods produced statistically comparable results in terms of success and prediction rates. However, with the Area Under the receiver operator Curve (AUC), statistical models are more accurate (88.7 and 84.6% for LR and WofE, respectively) than the PB model (only 69.8%). Nevertheless, in such data-sparse situation, the usual approaches for validation, i.e. comparing observed with predicted data, are insufficient, formal uncertainty analysis (UA) is a means for evaluating the validity and reliability of the model. We then refitted the PB model using a stochastic modification of the infinite slope stability model input scheme using Monte Carlo (MC) method backed with sensitivity analysis (SA). For statistical models, we used an informal Student t-test for estimating the certainty of the predicted probability (PP) at each location. Both modelling outputs independently show a high validity; and whereas the level of confidence in LR and WofE models remained the same after performance re-evaluation, the accuracy of the PB model showed an improvement (AUC = 72%). This result is reasonable and provides a further validation of PB model. So, in urban slope analysis, where PB diagnostic is necessary, statistical and PB modelling may play equally supportive roles in landslide hazard assessment. Numéro de notice : A2019-219 Affiliation des auteurs : non IGN Thématique : GEOMATIQUE Nature : Article nature-HAL : ArtAvecCL-RevueIntern DOI : 10.1080/10106049.2017.1357767 date de publication en ligne : 10/08/2017 En ligne : https://doi.org/10.1080/10106049.2017.1357767 Format de la ressource électronique : URL article Permalink : https://documentation.ensg.eu/index.php?lvl=notice_display&id=92737 [article]

Réservation

Réserver ce document

Exemplaires (1)

Code-barres	Cote	Support	Localisation	Section	Disponibilité
059-2019011	SL	Revue	Centre de documentation	Revues en salle	Disponible

Bayesian statistics and Monte Carlo methods / Karl Rudolf Koch in Journal of geodetic science, vol 8 n° 1 (January 2018)  

Public [article]  inJournal of geodetic science > vol 8 n° 1 (January 2018) . - pp 18 - 29 Titre : Bayesian statistics and Monte Carlo methods Type de document : Article/Communication Auteurs : Karl Rudolf Koch, Auteur Année de publication : 2018 Article en page(s) : pp 18 - 29 Note générale : Bibliographie Langues : Anglais (eng) Descripteur : [Vedettes matières IGN] Statistiques [Termes descripteurs IGN] matrice de covariance [Termes descripteurs IGN] méthode de Monte-Carlo [Termes descripteurs IGN] propagation d'erreur [Termes descripteurs IGN] théorème de Bayes [Termes descripteurs IGN] variable aléatoire [Termes descripteurs IGN] vecteur aléatoire multidimensionnel Résumé : (Auteur) The Bayesian approach allows an intuitive way to derive the methods of statistics. Probability is defined as a measure of the plausibility of statements or propositions. Three rules are sufficient to obtain the laws of probability. If the statements refer to the numerical values of variables, the so-called random variables, univariate and multivariate distributions follow. They lead to the point estimation by which unknown quantities, i.e. unknown parameters, are computed from measurements. The unknown parameters are random variables, they are fixed quantities in traditional statistics which is not founded on Bayes’ theorem. Bayesian statistics therefore recommends itself for Monte Carlo methods, which generate random variates from given distributions. Monte Carlo methods, of course, can also be applied in traditional statistics. The unknown parameters, are introduced as functions of the measurements, and the Monte Carlo methods give the covariance matrix and the expectation of these functions. A confidence region is derived where the unknown parameters are situated with a given probability. Following a method of traditional statistics, hypotheses are tested by determining whether a value for an unknown parameter lies inside or outside the confidence region. The error propagation of a random vector by the Monte Carlo methods is presented as an application. If the random vector results from a nonlinearly transformed vector, its covariance matrix and its expectation follow from the Monte Carlo estimate. This saves a considerable amount of derivatives to be computed, and errors of the linearization are avoided. The Monte Carlo method is therefore efficient. If the functions of the measurements are given by a sum of two or more random vectors with different multivariate distributions, the resulting distribution is generally not known. The Monte Carlo methods are then needed to obtain the covariance matrix and the expectation of the sum. Numéro de notice : A2018-613 Affiliation des auteurs : non IGN Thématique : POSITIONNEMENT Nature : Article nature-HAL : ArtAvecCL-RevueIntern DOI : 10.1515/jogs-2018-0003 date de publication en ligne : 02/03/2018 En ligne : https://doi.org/10.1515/jogs-2018-0003 Format de la ressource électronique : URL article Permalink : https://documentation.ensg.eu/index.php?lvl=notice_display&id=92646 [article]

Bayesian essentials with R / Jean-Michel Marin (2014)  

Public Titre : Bayesian essentials with R Type de document : Monographie Auteurs : Jean-Michel Marin, Auteur ; Christian P. Robert, Auteur Editeur : Berlin, Heidelberg, Vienne, New York, ... : Springer Année de publication : 2014 Importance : 296 p. Format : 16 x 24 cm ISBN/ISSN/EAN : 978-1-4614-8687-9 Note générale : bibliographie Langues : Anglais (eng) Descripteur : [Vedettes matières IGN] Statistiques [Termes descripteurs IGN] algorithme de Gibbs [Termes descripteurs IGN] algorithme de Métropolis-Hastings [Termes descripteurs IGN] classification bayesienne [Termes descripteurs IGN] méthode de Monte-Carlo par chaînes de Markov [Termes descripteurs IGN] modèle linéaire [Termes descripteurs IGN] problème de Dirichlet [Termes descripteurs IGN] R (langage) [Termes descripteurs IGN] régression linéaire [Termes descripteurs IGN] série temporelle [Termes descripteurs IGN] théorème de Bayes Résumé : (éditeur) This Bayesian modeling book provides a self-contained entry to computational Bayesian statistics. Focusing on the most standard statistical models and backed up by real datasets and an all-inclusive R (CRAN) package called bayess, the book provides an operational methodology for conducting Bayesian inference, rather than focusing on its theoretical and philosophical justifications. Readers are empowered to participate in the real-life data analysis situations depicted here from the beginning. The stakes are high and the reader determines the outcome. Special attention is paid to the derivation of prior distributions in each case and specific reference solutions are given for each of the models. Similarly, computational details are worked out to lead the reader towards an effective programming of the methods given in the book. In particular, all R codes are discussed with enough detail to make them readily understandable and expandable. This works in conjunction with the bayess package. Bayesian Essentials with R can be used as a textbook at both undergraduate and graduate levels, as exemplified by courses given at Université Paris Dauphine (France), University of Canterbury (New Zealand), and University of British Columbia (Canada). It is particularly useful with students in professional degree programs and scientists to analyze data the Bayesian way. The text will also enhance introductory courses on Bayesian statistics. Prerequisites for the book are an undergraduate background in probability and statistics, if not in Bayesian statistics. A strength of the text is the noteworthy emphasis on the role of models in statistical analysis. Note de contenu : 1- User’s Manual 2- Normal Models 3- Regression and Variable Selection 4- Generalized Linear Models 5- Capture–Recapture Experiments 6- Mixture Models 7- Time Series 8- Image Analysis Numéro de notice : 25759 Affiliation des auteurs : non IGN Thématique : MATHEMATIQUE Nature : Monographie En ligne : https://link.springer.com/book/10.1007%2F978-1-4614-8687-9#toc Format de la ressource électronique : URL Permalink : https://documentation.ensg.eu/index.php?lvl=notice_display&id=94954

Bayesian and frequentist regression methods / Jon Wakefield (2013)  

Permalink

Probabilités et statistiques, version préliminaire / Patrick Sillard (2000)

Permalink

Segmentierung und Interpretation digitaler Bilder mit Markoff-Zufallsfeldern / J. Klonowski (1999)

Permalink

Kontextsensitive Bildinterpretation mit Markoff-Zufallsfeldern / M. Köster (1995)

Permalink