Centre de documentation
scientifique

Spectral variability in hyperspectral unmixing : Multiscale, tensor, and neural network-based approaches / Ricardo Augusto Borsoi (2021)  

Public Titre : Spectral variability in hyperspectral unmixing : Multiscale, tensor, and neural network-based approaches Type de document : Thèse/HDR Auteurs : Ricardo Augusto Borsoi, Auteur ; Cédric Richard, Directeur de thèse ; José Carlos Moreira Bermudez, Directeur de thèse Editeur : Nice : Université Côte d'Azur Année de publication : 2021 Importance : 187 p. Format : 21 x 30 cm Note générale : bibliographie Thèse présentée en vue de l'obtention du grade de docteur science pour l’ingénieur de l’Université Côte d'Azur Langues : Anglais (eng) Descripteur : [Vedettes matières IGN] Traitement d'image optique [Termes IGN] analyse de mélange spectral d’extrémités multiples [Termes IGN] analyse linéaire des mélanges spectraux [Termes IGN] image hyperspectrale [Termes IGN] image multitemporelle [Termes IGN] réseau antagoniste génératif [Termes IGN] signature spectrale [Termes IGN] tenseur Index. décimale : THESE Thèses et HDR Résumé : (auteur) The spectral signatures of the materials contained in hyperspectral images, also called endmembers (EMs), can be significantly affected by variations in atmospheric, illumination or environmental conditions typically occurring within an image. Traditional spectral unmixing (SU) algorithms neglect the spectral variability of the endmembers, what propagates significant mismodeling errors throughout the whole unmixing process and compromises the quality of the estimated abundances. Therefore, significant effort have been recently dedicated to mitigate the effects of spectral variability in SU. However, many challenges still remain in how to best explore a priori information about the problem in order to improve the quality, the robustness and the efficiency of SU algorithms that account for spectral variability. In this thesis, new strategies are developed to address spectral variability in SU. First, an (over)-segmentation-based multiscale regularization strategy is proposed to explore spatial information about the abundance maps more effectively. New algorithms are then proposed for both semi-supervised and blind SU, leading to improved abundance reconstruction performance at a small computational complexity. Afterwards, three new models are proposed to represent spectral variability of the EMs in SU, using parametric, tensor, and neural network-based representations for EM spectra at each image pixel. The parametric model introduces pixel-dependent scaling factors over a reference EM matrix to model arbitrary spectral variability, while the tensor-based representation allows one to exploit the high-dimensional nature of the data by means of its underlying low-rank structure. Generative neural networks (such as variational autoencoders or generative adversarial networks) finally allow one to model the low-dimensional manifold of the spectral signatures of the materials more effectively. The proposed models are used to devise three new blind SU algorithms, and to perform data augmentation in library-based SU. Finally, we provide a brief overview of work which extends the proposed strategies to new problems in SU and in hyperspectral image analysis. This includes the use of the multiscale abundance regularization in nonlinear SU, modeling spectral variability and accounting for sudden changes when performing SU and change detection of multitemporal hyperspectral images, and also accounting for spectral variability and changes in the multimodal (i.e., hyperspectral and multispectral) image fusion problem. Note de contenu : 1- Introduction 2- Origin of linear mixing model spectral variability in hyperspectral images 3- A ultiscale spatial regularization for fast unmixing with spectral librairies 4- A data dependent multiscale model for spectral unmixing with specral variability 5- Generalized linear mixing model accounting for endmember variability 6- Low-rank tensor modeling for spectral unmixing accounting for spectral variability 7- Deep generative endmembers modeling: An application to unsupervised spectral unmixing 8- Deep generative models for library augmentation in multiple endmember spectral mixture analysis 9- And now for something different... 10- Conclusions Numéro de notice : 28487 Affiliation des auteurs : non IGN Thématique : IMAGERIE Nature : Thèse française Note de thèse : thèse de Doctorat : Sciences pour l'Ingénieur : Côte d'Azur : 2021 Organisme de stage : Laboratoire J.-L. Lagrange, Observatoire de la Côte d’Azur DOI : sans En ligne : https://tel.archives-ouvertes.fr/tel-03253631/document Format de la ressource électronique : URL Permalink : https://documentation.ensg.eu/index.php?lvl=notice_display&id=99188