Talks given at NuHAG events

Feature Preserving Image Denoising using Inter-and-intrascale Dependencies of Wavelet Coefficients

  Serena Papi
    Department of mathematics, University of Bologna

  given at  strobl07 (17.06.07)
  id:  728
  length:  min
  status:  accepted
  type:  poster


%\title{Edge Preserving Denoising via Anisotropic Wavelet Thresholding}
\title{Feature Preserving Image Denoising using Inter-and-intrascale Dependencies of Wavelet Coefficients}

\author{D.Lazzaro,\; L.B.Montefusco, \; S.Papi}


A new wavelet shrinkage algorithm for edge preserving image denoising is proposed. It represents an improvement of a previously
presented thresholding method \cite{MONTE} that adds to the denoising process the construction of the map of image edges.
The map is obtained by taking into consideration both space-scale dependencies and interscale correlations between corresponding coefficients
of the dyadic Mallat's wavelet decomposition of the image.
The knowledge of the location and the orientation of the edges allows us to perform an anisotropic shrinkage by killing the gradient components
along the edges while keeping or only slightly shrinking the orthogonal components. The non-edge coefficients are then shrinked using a
correlated Perona-Malik rule.
The resulting anisotropic wavelet thresholding scheme succeeds in producing a denoised version of the original image that still presents
well-defined edges, while the noise components are removed.


\bibitem{MONTE} D. Lazzaro, L.B.Montefusco, Edge-Preserving Wavelet Thresholding for Image Denoising,
J. Computat. Appl. Mat. (2006), doi:10.1016/


Enter here the CODE for editing this talk:
If you have forgotten the CODE for your talk click here to send an email to the Webmaster!
NOTICE: In [EDIT-MODUS] you can also UPLOAD a presentation"