We propose a soft thresholding approach to the minimum description length wavelet denoising. Our method is based on combining two-part coding with normalized maximum likelihood universal models to give a soft thresholding denoising criterion. Experiments with the
proposed MDL soft thresholding method indicate that our denoising criterion leads to fairly similar performance as with the well-known VisuShrink method.
OJANEN Janne;
HEIKKONEN Jukka;
2009-01-14
Tampere International Center for Signal Processing
JRC49470
http://sp.cs.tut.fi/WITMSE08/Proceedings/index.html,
http://sp.cs.tut.fi/WITMSE08/Proceedings/InvitedPapers/invited_Ojanen.pdf,
https://publications.jrc.ec.europa.eu/repository/handle/JRC49470,
This document is only visible at the Commission level.
You are not authorized to publish or distribute it outside the European Commission.
This is a public document. You can share this publication.
Additional supporting files
| File name | Description | File type | |
|---|
