Adaptive Structured Block Sparsity Via Dyadic Partitioning - Archive ouverte HAL Access content directly
Conference Papers Year : 2011

Adaptive Structured Block Sparsity Via Dyadic Partitioning

Abstract

This paper proposes a novel method to adapt the block-sparsity structure to the observed noisy data. Towards this goal, the Stein risk estimator framework is exploited, and the block-sparsity is dyadically organized in a tree. The adaptation of the sparsity structure is obtained by finding the best recursive dyadic partition, whose terminal nodes (leaves) are the blocks, that minimizes a data-driven estimator of the risk. Our main contributions are (i) analytical expression of the risk; (ii) a novel estimator of the risk; (iii) a fast algorithm that yields the best partition. Numerical results on wavelet-domain denoising of synthetic and natural images illustrate the improvement brought by our adaptive approach.
Fichier principal
Vignette du fichier
PeyreFadili-Eusipco11.pdf (1017.29 Ko) Télécharger le fichier
Origin : Files produced by the author(s)
Loading...

Dates and versions

hal-00597772 , version 1 (01-06-2011)

Identifiers

  • HAL Id : hal-00597772 , version 1

Cite

Gabriel Peyré, Jalal M. Fadili, Christophe Chesneau. Adaptive Structured Block Sparsity Via Dyadic Partitioning. EUSIPCO 2011, Aug 2011, Barcolona, Spain. pp.1455-1459. ⟨hal-00597772⟩
270 View
125 Download

Share

Gmail Facebook X LinkedIn More