Improving Angular Error by Near Circular Operator Design

Research output: Contribution to journalArticle

5 Citations (Scopus)

Abstract

In image filtering, the circularity of an operator is an important factor affecting its accuracy. When step edge orientation is estimated in a square neighbourhood, the use of standard methods can result in a detected orientation error of up to 6.6% [2]. Circular differential edge operators are effective in minimising this angular error and may in fact reduce it to zero for all orientations [2]. The principles of circularity [2] and scale (see, for example, [4]) are amongst the principal considerations when designing low-level image processing operators. When coupled with the task of designing optimal discrete Gaussian operators [1], such considerations become both particularly relevant and challenging. In this paper, we show how the adoption of a finite-element-based approach allows us to formulate a design procedure that can embrace all three aspects: circularity, scale and Gaussian approximation. Via the use of edge sensitivity analysis, we show that such a design procedure can significantly improve detected edge orientation over a full range of orientations and displacements compared with standard operators.
LanguageEnglish
Pages169-172
JournalPattern Recognition
Volume37
Issue number1
DOIs
Publication statusPublished - Jan 2004

Fingerprint

Sensitivity analysis
Image processing

Cite this

@article{ff404cb3fe70409e9b3e1fb30eb22b8b,
title = "Improving Angular Error by Near Circular Operator Design",
abstract = "In image filtering, the circularity of an operator is an important factor affecting its accuracy. When step edge orientation is estimated in a square neighbourhood, the use of standard methods can result in a detected orientation error of up to 6.6{\%} [2]. Circular differential edge operators are effective in minimising this angular error and may in fact reduce it to zero for all orientations [2]. The principles of circularity [2] and scale (see, for example, [4]) are amongst the principal considerations when designing low-level image processing operators. When coupled with the task of designing optimal discrete Gaussian operators [1], such considerations become both particularly relevant and challenging. In this paper, we show how the adoption of a finite-element-based approach allows us to formulate a design procedure that can embrace all three aspects: circularity, scale and Gaussian approximation. Via the use of edge sensitivity analysis, we show that such a design procedure can significantly improve detected edge orientation over a full range of orientations and displacements compared with standard operators.",
author = "BW Scotney and SA Coleman and MG Herron",
year = "2004",
month = "1",
doi = "10.1016/j.patcog.2003.07.001",
language = "English",
volume = "37",
pages = "169--172",
journal = "Pattern Recognition",
issn = "0031-3203",
publisher = "Elsevier",
number = "1",

}

Improving Angular Error by Near Circular Operator Design. / Scotney, BW; Coleman, SA; Herron, MG.

In: Pattern Recognition, Vol. 37, No. 1, 01.2004, p. 169-172.

Research output: Contribution to journalArticle

TY - JOUR

T1 - Improving Angular Error by Near Circular Operator Design

AU - Scotney, BW

AU - Coleman, SA

AU - Herron, MG

PY - 2004/1

Y1 - 2004/1

N2 - In image filtering, the circularity of an operator is an important factor affecting its accuracy. When step edge orientation is estimated in a square neighbourhood, the use of standard methods can result in a detected orientation error of up to 6.6% [2]. Circular differential edge operators are effective in minimising this angular error and may in fact reduce it to zero for all orientations [2]. The principles of circularity [2] and scale (see, for example, [4]) are amongst the principal considerations when designing low-level image processing operators. When coupled with the task of designing optimal discrete Gaussian operators [1], such considerations become both particularly relevant and challenging. In this paper, we show how the adoption of a finite-element-based approach allows us to formulate a design procedure that can embrace all three aspects: circularity, scale and Gaussian approximation. Via the use of edge sensitivity analysis, we show that such a design procedure can significantly improve detected edge orientation over a full range of orientations and displacements compared with standard operators.

AB - In image filtering, the circularity of an operator is an important factor affecting its accuracy. When step edge orientation is estimated in a square neighbourhood, the use of standard methods can result in a detected orientation error of up to 6.6% [2]. Circular differential edge operators are effective in minimising this angular error and may in fact reduce it to zero for all orientations [2]. The principles of circularity [2] and scale (see, for example, [4]) are amongst the principal considerations when designing low-level image processing operators. When coupled with the task of designing optimal discrete Gaussian operators [1], such considerations become both particularly relevant and challenging. In this paper, we show how the adoption of a finite-element-based approach allows us to formulate a design procedure that can embrace all three aspects: circularity, scale and Gaussian approximation. Via the use of edge sensitivity analysis, we show that such a design procedure can significantly improve detected edge orientation over a full range of orientations and displacements compared with standard operators.

UR - http://www.elsevier.com/locate/patcog

U2 - 10.1016/j.patcog.2003.07.001

DO - 10.1016/j.patcog.2003.07.001

M3 - Article

VL - 37

SP - 169

EP - 172

JO - Pattern Recognition

T2 - Pattern Recognition

JF - Pattern Recognition

SN - 0031-3203

IS - 1

ER -