Synthesis of orthogonal Roesser model for two-dimensional FIR filters

Author

Wirski, Robert T.

Author_Institution

Fac. of Electron. & Comput. Sci., Koszalin Univ. of Technol., Koszalin, Poland

fYear

2010

fDate

17-20 Oct. 2010

Firstpage

616

Lastpage

620

Abstract

In the paper, a novel synthesis technique of an orthogonal Roesser model for two-dimensional FIR filters is presented. It is based on the factorization of a transfer function coefficients matrix. Such an approach leads to synthesis of a cascade connection of one-dimensional systems. The numerical reliability of the technique is illustrated by examples of image convolution Gaussian filter and minimax design of a low-pass FIR system.

Keywords

FIR filters; Gaussian processes; convolution; image processing; low-pass filters; matrix decomposition; minimax techniques; transfer function matrices; image convolution Gaussian filter; low-pass FIR filter; matrix factorization; minimax design; numerical reliability; one-dimensional system; orthogonal Roesser model; transfer function coefficient matrix; two-dimensional FIR filter; Finite impulse response filter; Mathematical model; Numerical models; Polynomials; Scattering; Transfer functions; 2-D systems; FIR; orthogonal systems; state-space equations; synthesis;

fLanguage

English

Publisher

ieee

Conference_Titel

Information Theory and its Applications (ISITA), 2010 International Symposium on

Conference_Location

Taichung

Print_ISBN

978-1-4244-6016-8

Electronic_ISBN

978-1-4244-6017-5

Type

conf

DOI

10.1109/ISITA.2010.5649701

Filename

5649701