Wavelet deconvolution with noisy eigenvalues
Identifieur interne : 009250 ( Main/Exploration ); précédent : 009249; suivant : 009251Wavelet deconvolution with noisy eigenvalues
Auteurs : Laurent Cavalier [France] ; Marc Raimondo [Australie]Source :
- IEEE transactions on signal processing [ 1053-587X ] ; 2007.
Descripteurs français
- Pascal (Inist)
- Déconvolution, Valeur propre, Transformation ondelette, Reconstruction signal, Restauration signal, Image bruitée, Image floue, Evaluation performance, Approximation linéaire, Restauration image, Convolution, Signal entrée, Système linéaire, Estimation adaptative, Décomposition valeur singulière, Distribution Wigner Ville, Traitement signal, Traitement image.
English descriptors
- KwdEn :
- Adaptive estimation, Blurred image, Convolution, Deconvolution, Eigenvalue, Image processing, Image restoration, Input signal, Linear approximation, Linear system, Noisy image, Performance evaluation, Signal processing, Signal reconstruction, Signal restoration, Singular value decomposition, Wavelet transformation, Wigner Ville distribution.
Abstract
Over the last decade, there has been a lot of interest in wavelet-vaguelette methods for the recovery of noisy signals or images in motion blur. Nonlinear wavelet estimators are known to have good adaptive properties and to outperform linear approximations over a wide range of signals and images, see e.g., the recent WaveD method of Johnstone, Kerkyacharian, Picard, and Raimondo (2004) or the ForWarD method of Neelamani, Choi, and Baraniuk (2004) and also Fan and Koo (2002) in the density setting. In the deblurring setting, wavelet-vaguelette methods rely on the complete knowledge of a convolution operator's eigenvalues. This is an unlikely situation in practice, however. A more realistic scenario, such as would arise when passing the Fourier basis as an input signal through a linear-time-invariant system, is to imagine that one also observes a set of noisy eigenvalues. In this paper, we define a version of the WaveD estimator which is near-optimal when used with noisy eigenvalues. A key feature of our method includes a data-driven method for choosing the fine resolution level in WaveD estimation. Asymptotic theory is illustrated with a wide range of finite sample examples.
Affiliations:
- Australie, France
- Nouvelle-Galles du Sud, Provence-Alpes-Côte d'Azur
- Marseille, Sydney
- Université de Sydney
Links toward previous steps (curation, corpus...)
- to stream PascalFrancis, to step Corpus: 003C06
- to stream PascalFrancis, to step Curation: 002482
- to stream PascalFrancis, to step Checkpoint: 003512
- to stream Main, to step Merge: 009C24
- to stream Main, to step Curation: 009250
Le document en format XML
<record><TEI><teiHeader><fileDesc><titleStmt><title xml:lang="en" level="a">Wavelet deconvolution with noisy eigenvalues</title><author><name sortKey="Cavalier, Laurent" sort="Cavalier, Laurent" uniqKey="Cavalier L" first="Laurent" last="Cavalier">Laurent Cavalier</name><affiliation wicri:level="3"><inist:fA14 i1="01"><s1>Université Aix-Marseille 1, CMI</s1><s2>13453 Marseille</s2><s3>FRA</s3><sZ>1 aut.</sZ></inist:fA14><country>France</country><placeName><region type="region" nuts="2">Provence-Alpes-Côte d'Azur</region><settlement type="city">Marseille</settlement></placeName></affiliation></author><author><name sortKey="Raimondo, Marc" sort="Raimondo, Marc" uniqKey="Raimondo M" first="Marc" last="Raimondo">Marc Raimondo</name><affiliation wicri:level="4"><inist:fA14 i1="02"><s1>School of Mathematics and Statistics, University of Sydney</s1><s2>NSW 2006</s2><s3>AUS</s3><sZ>2 aut.</sZ></inist:fA14><country>Australie</country><placeName><settlement type="city">Sydney</settlement><region type="état">Nouvelle-Galles du Sud</region></placeName><orgName type="university">Université de Sydney</orgName></affiliation></author></titleStmt><publicationStmt><idno type="wicri:source">INIST</idno><idno type="inist">07-0268082</idno><date when="2007">2007</date><idno type="stanalyst">PASCAL 07-0268082 INIST</idno><idno type="RBID">Pascal:07-0268082</idno><idno type="wicri:Area/PascalFrancis/Corpus">003C06</idno><idno type="wicri:Area/PascalFrancis/Curation">002482</idno><idno type="wicri:Area/PascalFrancis/Checkpoint">003512</idno><idno type="wicri:explorRef" wicri:stream="PascalFrancis" wicri:step="Checkpoint">003512</idno><idno type="wicri:doubleKey">1053-587X:2007:Cavalier L:wavelet:deconvolution:with</idno><idno type="wicri:Area/Main/Merge">009C24</idno><idno type="wicri:Area/Main/Curation">009250</idno><idno type="wicri:Area/Main/Exploration">009250</idno></publicationStmt><sourceDesc><biblStruct><analytic><title xml:lang="en" level="a">Wavelet deconvolution with noisy eigenvalues</title><author><name sortKey="Cavalier, Laurent" sort="Cavalier, Laurent" uniqKey="Cavalier L" first="Laurent" last="Cavalier">Laurent Cavalier</name><affiliation wicri:level="3"><inist:fA14 i1="01"><s1>Université Aix-Marseille 1, CMI</s1><s2>13453 Marseille</s2><s3>FRA</s3><sZ>1 aut.</sZ></inist:fA14><country>France</country><placeName><region type="region" nuts="2">Provence-Alpes-Côte d'Azur</region><settlement type="city">Marseille</settlement></placeName></affiliation></author><author><name sortKey="Raimondo, Marc" sort="Raimondo, Marc" uniqKey="Raimondo M" first="Marc" last="Raimondo">Marc Raimondo</name><affiliation wicri:level="4"><inist:fA14 i1="02"><s1>School of Mathematics and Statistics, University of Sydney</s1><s2>NSW 2006</s2><s3>AUS</s3><sZ>2 aut.</sZ></inist:fA14><country>Australie</country><placeName><settlement type="city">Sydney</settlement><region type="état">Nouvelle-Galles du Sud</region></placeName><orgName type="university">Université de Sydney</orgName></affiliation></author></analytic><series><title level="j" type="main">IEEE transactions on signal processing</title><title level="j" type="abbreviated">IEEE trans. signal process.</title><idno type="ISSN">1053-587X</idno><imprint><date when="2007">2007</date></imprint></series></biblStruct></sourceDesc><seriesStmt><title level="j" type="main">IEEE transactions on signal processing</title><title level="j" type="abbreviated">IEEE trans. signal process.</title><idno type="ISSN">1053-587X</idno></seriesStmt></fileDesc><profileDesc><textClass><keywords scheme="KwdEn" xml:lang="en"><term>Adaptive estimation</term><term>Blurred image</term><term>Convolution</term><term>Deconvolution</term><term>Eigenvalue</term><term>Image processing</term><term>Image restoration</term><term>Input signal</term><term>Linear approximation</term><term>Linear system</term><term>Noisy image</term><term>Performance evaluation</term><term>Signal processing</term><term>Signal reconstruction</term><term>Signal restoration</term><term>Singular value decomposition</term><term>Wavelet transformation</term><term>Wigner Ville distribution</term></keywords><keywords scheme="Pascal" xml:lang="fr"><term>Déconvolution</term><term>Valeur propre</term><term>Transformation ondelette</term><term>Reconstruction signal</term><term>Restauration signal</term><term>Image bruitée</term><term>Image floue</term><term>Evaluation performance</term><term>Approximation linéaire</term><term>Restauration image</term><term>Convolution</term><term>Signal entrée</term><term>Système linéaire</term><term>Estimation adaptative</term><term>Décomposition valeur singulière</term><term>Distribution Wigner Ville</term><term>Traitement signal</term><term>Traitement image</term></keywords></textClass></profileDesc></teiHeader><front><div type="abstract" xml:lang="en">Over the last decade, there has been a lot of interest in wavelet-vaguelette methods for the recovery of noisy signals or images in motion blur. Nonlinear wavelet estimators are known to have good adaptive properties and to outperform linear approximations over a wide range of signals and images, see e.g., the recent WaveD method of Johnstone, Kerkyacharian, Picard, and Raimondo (2004) or the ForWarD method of Neelamani, Choi, and Baraniuk (2004) and also Fan and Koo (2002) in the density setting. In the deblurring setting, wavelet-vaguelette methods rely on the complete knowledge of a convolution operator's eigenvalues. This is an unlikely situation in practice, however. A more realistic scenario, such as would arise when passing the Fourier basis as an input signal through a linear-time-invariant system, is to imagine that one also observes a set of noisy eigenvalues. In this paper, we define a version of the WaveD estimator which is near-optimal when used with noisy eigenvalues. A key feature of our method includes a data-driven method for choosing the fine resolution level in WaveD estimation. Asymptotic theory is illustrated with a wide range of finite sample examples.</div></front></TEI><affiliations><list><country><li>Australie</li><li>France</li></country><region><li>Nouvelle-Galles du Sud</li><li>Provence-Alpes-Côte d'Azur</li></region><settlement><li>Marseille</li><li>Sydney</li></settlement><orgName><li>Université de Sydney</li></orgName></list><tree><country name="France"><region name="Provence-Alpes-Côte d'Azur"><name sortKey="Cavalier, Laurent" sort="Cavalier, Laurent" uniqKey="Cavalier L" first="Laurent" last="Cavalier">Laurent Cavalier</name></region></country><country name="Australie"><region name="Nouvelle-Galles du Sud"><name sortKey="Raimondo, Marc" sort="Raimondo, Marc" uniqKey="Raimondo M" first="Marc" last="Raimondo">Marc Raimondo</name></region></country></tree></affiliations></record>Pour manipuler ce document sous Unix (Dilib)
EXPLOR_STEP=$WICRI_ROOT/Wicri/Asie/explor/AustralieFrV1/Data/Main/Exploration
HfdSelect -h $EXPLOR_STEP/biblio.hfd -nk 009250 | SxmlIndent | more
Ou
HfdSelect -h $EXPLOR_AREA/Data/Main/Exploration/biblio.hfd -nk 009250 | SxmlIndent | more
Pour mettre un lien sur cette page dans le réseau Wicri
{{Explor lien
|wiki= Wicri/Asie
|area= AustralieFrV1
|flux= Main
|étape= Exploration
|type= RBID
|clé= Pascal:07-0268082
|texte= Wavelet deconvolution with noisy eigenvalues
}}
| This area was generated with Dilib version V0.6.33. |  |