A variational Bayesian approximation approach via a sparsity enforcing prior in acoustic imaging

Ning Chu; Ali Mohammad-Djafari; Nicolas Gac; José Picheral

doi:10.1109/WIO.2014.6933297

A variational Bayesian approximation approach via a sparsity enforcing prior in acoustic imaging

Ning Chu ^{1, 2} Ali Mohammad-Djafari ¹ Nicolas Gac ¹ José Picheral ²

1 L2S - Laboratoire des signaux et systèmes

2 E3S - Supélec Sciences des Systèmes [Gif-sur-Yvette]

Abstract : Acoustic imaging is an advanced technique for acoustic source localization and power reconstruction from limited noisy measurements at microphone sensors. To solve this ill-posed inverse problem, the Bayesian inference methods using proper prior knowledge have been widely investigated. In this paper, we propose to use a hierarchical Variational Bayesian Approximation for the robust acoustic imaging. And we explore the Student's-t priors with heavy tails to enforce source sparsity and non-Gaussian noises, so that we can achieve the super spatial resolution and wide dynamic range of source powers. In addition, proposed approach is validated by simulations and real data from wind tunnel in automobile industry.

Keywords : Variational Bayesian Approximation Acoustic imaging non-Gaussian noises Student's-t prior

Type de document :

Communication dans un congrès

WIO 2014, Jul 2014, Neuchâtel, Switzerland. Proceedings of the 2014 13th Workshop on Information Optics, pp.1 - 4, 2014, 〈10.1109/WIO.2014.6933297〉

Domaine :

Sciences de l'ingénieur [physics] / Traitement du signal et de l'image

Liste complète des métadonnées

Littérature citée [11 références] Voir Masquer Télécharger

https://hal-supelec.archives-ouvertes.fr/hal-01103751
Contributeur : Alexandra Siebert <>
Soumis le : jeudi 15 janvier 2015 - 12:45:36
Dernière modification le : dimanche 16 septembre 2018 - 22:06:06
Document(s) archivé(s) le : jeudi 16 avril 2015 - 10:35:52

WIO2014.pdf

Fichiers produits par l'(les) auteur(s)

A variational Bayesian approximation approach via a sparsity enforcing prior in acoustic imaging

Fichier

Identifiants

Collections

Citation

Exporter

Métriques

373

156

Contact

A variational Bayesian approximation approach via a sparsity enforcing prior in acoustic imaging

Fichier

Identifiants

Collections

Citation

Exporter

Partager

Métriques

373

156

Contact