Summary: To appear: 3rd IEEE Int'l Conf on Image Processing. Lausanne, Switzerland. September 1996.
NOISE REMOVAL VIA BAYESIAN WAVELET CORING
Eero P. Simoncelli
Computer and Information Science Dept.
University of Pennsylvania
Philadelphia, PA 19104
Edward H. Adelson
Brain and Cognitive Science Dept.
Massachusetts Institute of Technology
Cambridge, MA 02139
The classical solution to the noise removal problem is
the Wiener lter, which utilizes the second-order statis-
tics of the Fourier decomposition. Subband decomposi-
tions of natural images have signicantly non-Gaussian
higher-order point statistics; these statistics capture im-
age properties that elude Fourier-based techniques. We
develop a Bayesian estimator that is a natural exten-
sion of the Wiener solution, and that exploits these
higher-order statistics. The resulting nonlinear esti-
mator performs a \coring" operation. We provide a