Physics-driven inverse problems made tractable with cosparse regularization

Abstract : Sparse data models are powerful tools for solving ill-posed inverse problems. We present a regularization framework based on the sparse synthesis and sparse analysis models for problems governed by linear partial differential equations. Although nominally equivalent, we show that the two models differ substantially from a computational perspective: unlike the sparse synthesis model, its analysis counterpart has much better scaling capabilities and can indeed be faster when more measurement data is available. Our findings are illustrated on two examples, sound source localization and brain source localization, which also serve as showcases for the regularization framework. To address this type of inverse problems, we develop a specially tailored convex optimization algorithm based on the Alternating Direction Method of Multipliers.
Type de document :
Article dans une revue
IEEE Transactions on Signal Processing, Institute of Electrical and Electronics Engineers, 2016, 64 (2), pp.335-348. 〈10.1109/TSP.2015.2480045〉
Liste complète des métadonnées

Littérature citée [57 références]  Voir  Masquer  Télécharger

https://hal.inria.fr/hal-01133087
Contributeur : Srdan Kitic <>
Soumis le : vendredi 21 août 2015 - 17:26:55
Dernière modification le : jeudi 15 novembre 2018 - 11:58:46
Document(s) archivé(s) le : mercredi 26 avril 2017 - 10:21:59

Fichiers

CosparsePhysics_double.pdf
Fichiers produits par l'(les) auteur(s)

Identifiants

Citation

Srđan Kitić, Laurent Albera, Nancy Bertin, Rémi Gribonval. Physics-driven inverse problems made tractable with cosparse regularization. IEEE Transactions on Signal Processing, Institute of Electrical and Electronics Engineers, 2016, 64 (2), pp.335-348. 〈10.1109/TSP.2015.2480045〉. 〈hal-01133087v3〉

Partager

Métriques

Consultations de la notice

1552

Téléchargements de fichiers

6158