Generalised class of nonlinear-type hybrid filters

L. Khriji, M. Gabbouj

Research output: Contribution to journalArticle

4 Citations (Scopus)

Abstract

A new class of nonlinear filters called the generalised directional distance rational hybrid filters (GDDRHF) for multidimensional signal processing is described. The application at hand is colour image filtering problems. The GDDRHF filter is a two-stage filter, which exploits the features of a new directional distance filter, where the temporal information related to the central sample is expressed by its weight, and those of the rational operator. This weight is incorporated to a sum of vector distances and vector angles. It is shown that the new GDDRHF outperforms a number of widely known nonlinear filters with respect to all criteria used.

Original languageEnglish
Pages (from-to)1650-1651
Number of pages2
JournalElectronics Letters
Volume38
Issue number25
DOIs
Publication statusPublished - Dec 5 2002

Fingerprint

Signal processing
Color

ASJC Scopus subject areas

  • Electrical and Electronic Engineering

Cite this

Generalised class of nonlinear-type hybrid filters. / Khriji, L.; Gabbouj, M.

In: Electronics Letters, Vol. 38, No. 25, 05.12.2002, p. 1650-1651.

Research output: Contribution to journalArticle

Khriji, L. ; Gabbouj, M. / Generalised class of nonlinear-type hybrid filters. In: Electronics Letters. 2002 ; Vol. 38, No. 25. pp. 1650-1651.
@article{9d41a62c33a24004accd214ee259bbce,
title = "Generalised class of nonlinear-type hybrid filters",
abstract = "A new class of nonlinear filters called the generalised directional distance rational hybrid filters (GDDRHF) for multidimensional signal processing is described. The application at hand is colour image filtering problems. The GDDRHF filter is a two-stage filter, which exploits the features of a new directional distance filter, where the temporal information related to the central sample is expressed by its weight, and those of the rational operator. This weight is incorporated to a sum of vector distances and vector angles. It is shown that the new GDDRHF outperforms a number of widely known nonlinear filters with respect to all criteria used.",
author = "L. Khriji and M. Gabbouj",
year = "2002",
month = "12",
day = "5",
doi = "10.1049/el:20021120",
language = "English",
volume = "38",
pages = "1650--1651",
journal = "Electronics Letters",
issn = "0013-5194",
publisher = "Institution of Engineering and Technology",
number = "25",

}

TY - JOUR

T1 - Generalised class of nonlinear-type hybrid filters

AU - Khriji, L.

AU - Gabbouj, M.

PY - 2002/12/5

Y1 - 2002/12/5

N2 - A new class of nonlinear filters called the generalised directional distance rational hybrid filters (GDDRHF) for multidimensional signal processing is described. The application at hand is colour image filtering problems. The GDDRHF filter is a two-stage filter, which exploits the features of a new directional distance filter, where the temporal information related to the central sample is expressed by its weight, and those of the rational operator. This weight is incorporated to a sum of vector distances and vector angles. It is shown that the new GDDRHF outperforms a number of widely known nonlinear filters with respect to all criteria used.

AB - A new class of nonlinear filters called the generalised directional distance rational hybrid filters (GDDRHF) for multidimensional signal processing is described. The application at hand is colour image filtering problems. The GDDRHF filter is a two-stage filter, which exploits the features of a new directional distance filter, where the temporal information related to the central sample is expressed by its weight, and those of the rational operator. This weight is incorporated to a sum of vector distances and vector angles. It is shown that the new GDDRHF outperforms a number of widely known nonlinear filters with respect to all criteria used.

UR - http://www.scopus.com/inward/record.url?scp=0037028240&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0037028240&partnerID=8YFLogxK

U2 - 10.1049/el:20021120

DO - 10.1049/el:20021120

M3 - Article

VL - 38

SP - 1650

EP - 1651

JO - Electronics Letters

JF - Electronics Letters

SN - 0013-5194

IS - 25

ER -