Afficher la notice abrégée

Image super-resolution with PCA Reduced generalized Gaussian mixture models

hal.structure.identifierInstitut de Mathématiques de Bordeaux [IMB]
hal.structure.identifierLaboratoire de l'intégration, du matériau au système [IMS]
dc.contributor.authorNGUYEN, Dang-Phuong-Lan
hal.structure.identifierTechnical University of Berlin / Technische Universität Berlin [TUB]
dc.contributor.authorHERTRICH, Johannes
hal.structure.identifierInstitut de Mathématiques de Bordeaux [IMB]
dc.contributor.authorAUJOL, Jean-Francois
hal.structure.identifierLaboratoire de l'intégration, du matériau au système [IMS]
dc.contributor.authorBERTHOUMIEU, Yannick
dc.date.accessioned2024-04-04T02:41:25Z
dc.date.available2024-04-04T02:41:25Z
dc.identifier.urihttps://oskar-bordeaux.fr/handle/20.500.12278/191150
dc.description.abstractEnSingle Image Super-Resolution algorithms based on patches have been noticed and widely used over the past decade. Recently, generalized Gaussian mixture models (GGMMs) have been shown to be a suitable tool for many image processing problems because of the flexible shape parameter. In this work, we first propose to use a joint GGMM learned from concatenated vectors of high-and low-resolution training patches. For each low-resolution patch, we compute the minimum mean square error (MMSE) estimator and generate the high-resolution image by averaging these estimates. We select the MMSE approach using GGMM as the method is invariant to affine contrast change and also invariant to a linear super-resolution operator. Unfortunately, the large dimension of the concatenated high-and low-resolution patches leads to instabilities and an intractable computational effort when estimating the parameters of the GGMM. Thus, we propose to combine a GGMM with a principal component analysis and derive an EM algorithm for estimating the parameters of the arising model. We demonstrate the performance of our model by numerical examples on synthetic and real images of materials' microstructures.
dc.language.isoen
dc.subject.enGeneralized Gaussian mixture model
dc.subject.enimage super-resolution
dc.subject.enhigh-dimensional data
dc.subject.endimensionality reduction
dc.subject.enPrincipal Component Analysis (PCA)
dc.title.enImage super-resolution with PCA Reduced generalized Gaussian mixture models
dc.typeDocument de travail - Pré-publication
dc.subject.halStatistiques [stat]/Machine Learning [stat.ML]
dc.subject.halInformatique [cs]/Apprentissage [cs.LG]
dc.subject.halMathématiques [math]/Statistiques [math.ST]
dc.subject.halSciences de l'ingénieur [physics]/Traitement du signal et de l'image
dc.subject.halInformatique [cs]/Traitement des images
bordeaux.hal.laboratoriesInstitut de Mathématiques de Bordeaux (IMB) - UMR 5251*
bordeaux.institutionUniversité de Bordeaux
bordeaux.institutionBordeaux INP
bordeaux.institutionCNRS
hal.identifierhal-03664839
hal.version1
hal.origin.linkhttps://hal.archives-ouvertes.fr//hal-03664839v1
bordeaux.COinSctx_ver=Z39.88-2004&rft_val_fmt=info:ofi/fmt:kev:mtx:journal&rft.au=NGUYEN,%20Dang-Phuong-Lan&HERTRICH,%20Johannes&AUJOL,%20Jean-Francois&BERTHOUMIEU,%20Yannick&rft.genre=preprint


Fichier(s) constituant ce document

FichiersTailleFormatVue

Il n'y a pas de fichiers associés à ce document.

Ce document figure dans la(les) collection(s) suivante(s)

Afficher la notice abrégée