A statistical extension of normalized convolution and its usage for image interpolation and filtering

Author

Muhlich, Matthias ; Mester, Rudolf

Author_Institution

Inst. for Appl. Phys., J.W. Goethe-Univ. Frankfurt, Frankfurt am Main, Germany

fYear

2004

fDate

6-10 Sept. 2004

Firstpage

489

Lastpage

492

Abstract

The natural characteristics of image signals and the statistics of measurement noise are decisive for designing optimal filter sets and optimal estimation methods in signal processing. Astonishingly, this principle has so far only partially found its way into the field of image sequence processing. We show how a Wiener-type MMSE optimization criterion for the resulting image signal, based on a simple covariance model of images or image sequences provides direct and intelligible solution for various, apparently different problems, such as error concealment, or adaption of filters to signal and noise statistics.

Keywords

Wiener filters; convolution; filtering theory; image sequences; interpolation; least mean squares methods; optimisation; statistical analysis; Wiener-type MMSE optimization criterion; covariance image model; image filtering; image interpolation; image sequence processing; image signal characteristics; measurement noise; noise statistics; normalized convolution; optimal estimation methods; optimal filter set design; signal processing; statistical extension; Abstracts; Image edge detection; Information filters; Noise;

fLanguage

English

Publisher

ieee

Conference_Titel

Signal Processing Conference, 2004 12th European

Conference_Location

Vienna

Print_ISBN

978-320-0001-65-7

Type

conf

Filename

7080015