Accelerated Dictionary Learning for Sparse Signal Representation

Fateme Ghayem; Mostafa Sadeghi; Massoud Babaie-Zadeh; Christian Jutten

Accelerated Dictionary Learning for Sparse Signal Representation

Fateme Ghayem ¹ Mostafa Sadeghi ¹ Massoud Babaie-Zadeh ¹ Christian Jutten ²

1 EE department of SUT [Tehran] - Department of Electrical Engineering [Tehran]

2 GIPSA-VIBS - VIBS

GIPSA-DIS - Département Images et Signal

Abstract : Learning sparsifying dictionaries from a set of training signals has been shown to have much better performance than pre-designed dictionaries in many signal processing tasks, including image enhancement. To this aim, numerous practical dictionary learning (DL) algorithms have been proposed over the last decade. This paper introduces an accelerated DL algorithm based on iterative proximal methods. The new algorithm efficiently utilizes the iterative nature of DL process, and uses accelerated schemes for updating dictionary and coefficient matrix. Our numerical experiments on dictionary recovery show that, compared with some well-known DL algorithms, our proposed one has a better convergence rate. It is also able to successfully recover underlying dictionaries for different sparsity and noise levels.

Keywords : proximal algorithms dictionary learning compressed sensing Sparse representation

Document type :

Conference papers

Domain :

Computer Science [cs] / Signal and Image Processing

Complete list of metadatas

https://hal.archives-ouvertes.fr/hal-01479437
Contributor : Christian Jutten <>
Submitted on : Tuesday, February 28, 2017 - 6:41:49 PM
Last modification on : Monday, July 8, 2019 - 3:09:25 PM
Long-term archiving on : Monday, May 29, 2017 - 4:24:08 PM

Fateme-LVA_ICA_final.pdf

Files produced by the author(s)

Accelerated Dictionary Learning for Sparse Signal Representation

File

Identifiers

Collections

Citation

Export

Metrics

464

260

Contact

Accelerated Dictionary Learning for Sparse Signal Representation

File

Identifiers

Collections

Citation

Export

Share

Metrics

464

260

Contact