APPROXIMATION OF 2-D FILTERS USING CANONIC SEPARABLE DENOMINATOR STATE-SPACE MODELS

被引：0

作者：

HINAMOTO, T ^{[1
]}

FAIRMAN, FW ^{[1
]}

MUNEYASU, M ^{[1
]}

MAEKAWA, S ^{[1
]}

机构：

[1] KOBE UNIV,DEPT SYST ENGN,KOBE 657,JAPAN

来源：

INTERNATIONAL JOURNAL OF SYSTEMS SCIENCE | 1985年 / 16卷 / 10期

关键词：

2-D FILTERS - CANONIC SEPARABLE DENOMINATOR STATE-SPACE MODELS;

D O I：

10.1080/00207728508926744

中图分类号：

TP [自动化技术、计算机技术];

学科分类号：

0812 ;

摘要：

A state-space model capable of representing 2-D transfer functions having separable denominators is introduced. The coordinates for the state space are chosen so that the number of adjustable scalar parameters in the model is minimized. This minimum-parameter model is used to approximate a 2-D filter by adjusting the model parameters so as to minimize a weighted sum of the squared difference between the 2-D filter and model impulse responses over the closed quarter-plane. This adjustment is done using the Davidon-Fletcher and Powell optimization algorithm. An example is given to illustrate the utility of the proposed method.

引用

页码：1207 / 1216

页数：10