Tuesday, 27 July 2004, 2pm:

<CENTER><B>Automatic Restoration of Visual Images
</B>

<P>
Javier Portilla<BR>
Universidad de Granada

</CENTER>


Gaussian scale mixtures (GSM) capture two basic properties of
the wavelet coefficients responding to natural images, namely
1) high kurtosis marginals, and 2) positive covariance between
neighbor coefficient amplitudes. These features are not shared
by Gaussian or lower kurtosis noise sources. Therefore, GSM
models provide a means to separate the noise from the signal
for an observed corrupted image. A local model consisting of
a GSM term plus Gaussian additive noise with arbitrary covariance
is used to estimate first the noise covariance at each wavelet
subband, applying a generalized expectation maximization algorithm.
Then the original wavelet coefficients are estimated from
the noisy observations using an efficient Bayes Least Squares
technique. Both steps are fully automatic.