University of Surrey

Test tubes in the lab Research in the ATI Dance Research

Learning Incoherent Dictionaries for Sparse Approximation Using Iterative Projections and Rotations

Barchiesi, D and Plumbley, MD (2013) Learning Incoherent Dictionaries for Sparse Approximation Using Iterative Projections and Rotations IEEE TRANSACTIONS ON SIGNAL PROCESSING, 61 (8). pp. 2055-2065.

[img]
Preview
Text
BarchiesiPlumbley13-tsp_accepted_notice.pdf - Accepted version Manuscript
Available under License : See the attached licence file.

Download (808kB) | Preview
[img]
Preview
PDF (licence)
SRI_deposit_agreement.pdf
Available under License : See the attached licence file.

Download (33kB) | Preview

Abstract

This article deals with learning dictionaries for sparse approximation whose atoms are both adapted to a training set of signals and mutually incoherent. To meet this objective, we employ a dictionary learning scheme consisting of sparse approximation followed by dictionary update and we add to the latter a decorrelation step in order to reach a target mutual coherence level. This step is accomplished by an iterative projection method complemented by a rotation of the dictionary. Experiments on musical audio data and a comparison with the method of optimal coherence-constrained directions (mocod) and the incoherent k-svd (ink-svd) illustrate that the proposed algorithm can learn dictionaries that exhibit a low mutual coherence while providing a sparse approximation with better signal-to-noise ratio (snr) than the benchmark techniques

Item Type: Article
Subjects : Electronic Engineering
Divisions : Faculty of Engineering and Physical Sciences > Electronic Engineering
Authors :
AuthorsEmailORCID
Barchiesi, DUNSPECIFIEDUNSPECIFIED
Plumbley, MDUNSPECIFIEDUNSPECIFIED
Date : 1 April 2013
Identification Number : https://doi.org/10.1109/TSP.2013.2245663
Copyright Disclaimer : © 2013 IEEE. Personal use of this material is permitted. Permission from IEEE must be obtained for all other uses, in any current or future media, including reprinting/republishing this material for advertising or promotional purposes, creating new collective works, for resale or redistribution to servers or lists, or reuse of any copyrighted component of this work in other works.
Uncontrolled Keywords : Science & Technology, Technology, Engineering, Electrical & Electronic, Engineering, ENGINEERING, ELECTRICAL & ELECTRONIC, Dictionary learning, iterative projections, mutual coherence, sparse approximation, MORPHOLOGICAL COMPONENT ANALYSIS, OVERCOMPLETE DICTIONARIES, MATRIX-FACTORIZATION, MATCHING PURSUITS, REPRESENTATIONS, ALGORITHMS, DESIGN, FRAMES
Related URLs :
Depositing User : Symplectic Elements
Date Deposited : 21 Sep 2016 09:20
Last Modified : 21 Sep 2016 09:20
URI: http://epubs.surrey.ac.uk/id/eprint/812258

Actions (login required)

View Item View Item

Downloads

Downloads per month over past year


Information about this web site

© The University of Surrey, Guildford, Surrey, GU2 7XH, United Kingdom.
+44 (0)1483 300800