Show simple item record

dc.contributor.authorDe Canditiis, Daniela
dc.contributor.authorVidakovic, Brani
dc.date.accessioned2008-12-10T16:34:22Z
dc.date.available2008-12-10T16:34:22Z
dc.date.issued2001
dc.identifier.urihttp://hdl.handle.net/1853/25935
dc.description.abstractIn this paper we propose a block shrinkage method in the wavelet domain for estimating an unknown function in the presence of Gaussian noise. This shrinkage utilizes an empirical Bayes, block-adaptive approach that accounts for the sparseness of the representation of the unknown function. The modeling is accomplished by using a mixture of two normal-inverse gamma (N I G) distributions as a joint prior on wavelet coefficients and noise variance in each block at a particular resolution level. This method results in explicit and fast rules. An automatic, level dependent choice for the prior hyperparameters is also suggested. Finally, the performance of the proposed method, BBS (Bayesian Block Shrinkage), is illustrated on the battery of standard test functions and compared to some standard wavelet-based denoising methods.en
dc.language.isoen_USen
dc.publisherGeorgia Institute of Technologyen
dc.relation.ispartofseriesBiomedical Engineering Technical Report ; G05/2001en
dc.subjectWavelet regressionen
dc.subjectShrinkageen
dc.subjectBayesian estimationen
dc.subjectNormal-inverse gamma priorsen
dc.titleWavelet Bayesian Block Shrinkage via Mixtures of Normal-Inverse Gamma Priorsen
dc.typeTechnical Reporten
dc.contributor.corporatenameGeorgia Institute of Technology
dc.contributor.corporatenameConsiglio nazionale delle ricerche (Italy). Istituto per Applicazioni della Matematica


Files in this item

Thumbnail

This item appears in the following Collection(s)

Show simple item record