Skip to Main content Skip to Navigation
Journal articles

Blind Source Separation with Outliers in Transformed Domains

Abstract : Blind Source Separation (BSS) methods are well suited for the analysis of multichannel data. In many applications, the observations are corrupted by an additional structured noise, which hinders most of the standard BSS techniques. In this article, we propose a novel BSS method able to jointly unmix the sources and separate the source contribution from the structured noise or outliers. This separation builds upon the difference of morphology between the components of interest, often encountered in imaging problems, by exploiting a sparse modeling of the components in two different domains. Numerical experiments highlight the robustness and precision of the proposed method in a wide variety of settings, including the full-rank regime.
Complete list of metadata

https://hal.archives-ouvertes.fr/hal-03177865
Contributor : Jerome Bobin <>
Submitted on : Tuesday, March 23, 2021 - 2:38:02 PM
Last modification on : Thursday, March 25, 2021 - 3:19:59 AM

File

Chenot18.pdf
Files produced by the author(s)

Identifiers

Citation

Cécile Chenot, Jerome Bobin. Blind Source Separation with Outliers in Transformed Domains. SIAM Journal on Imaging Sciences, Society for Industrial and Applied Mathematics, 2018, 11 (2), pp.1524-1559. ⟨10.1137/17m1133919⟩. ⟨hal-03177865⟩

Share

Metrics

Record views

25

Files downloads

52