Energy of asymmetry for accurate detection of global reflection axes

Dinggang Shen, Horace H S Ip, Eam Khwang Teoh

Research output: Contribution to journalArticle

14 Citations (Scopus)

Abstract

We define in this paper two energy terms, symmetric energy term and asymmetric energy term, which respectively correspond to the symmetric and asymmetric components of an object. The asymmetric energy term must be zero, if the studied object is invariant under a reflection about the x-axis. Accordingly, we formulate the problem of detecting reflectional symmetries as a problem of minimizing the asymmetric energy term. From the local minima of the asymmetric energy term, we can detect all the symmetric axes of any object. Since the asymmetric energy term is expressed as a summation of a set of generalized complex (GC) moments computed for an object, the proposed symmetry detection method is robust against both noise and slight deformation. We use the steepest descent technique to calculate the local minima of the asymmetric energy term, whose initialization is calculated from the most dominant GC moment. Experiments on typical logo images and human brain image have shown the effectiveness and the robustness of the proposed method. To our knowledge, this is the first theory on energy functions that describe the symmetric and asymmetric components of a 2D pattern.

Original languageEnglish
Pages (from-to)283-297
Number of pages15
JournalImage and Vision Computing
Volume19
Issue number5
DOIs
Publication statusPublished - 2001 Apr 1
Externally publishedYes

Fingerprint

Brain
Experiments

ASJC Scopus subject areas

  • Computer Vision and Pattern Recognition
  • Signal Processing
  • Electrical and Electronic Engineering

Cite this

Energy of asymmetry for accurate detection of global reflection axes. / Shen, Dinggang; Ip, Horace H S; Teoh, Eam Khwang.

In: Image and Vision Computing, Vol. 19, No. 5, 01.04.2001, p. 283-297.

Research output: Contribution to journalArticle

Shen, Dinggang ; Ip, Horace H S ; Teoh, Eam Khwang. / Energy of asymmetry for accurate detection of global reflection axes. In: Image and Vision Computing. 2001 ; Vol. 19, No. 5. pp. 283-297.
@article{fa224a929fb74fad9d05307405176a9f,
title = "Energy of asymmetry for accurate detection of global reflection axes",
abstract = "We define in this paper two energy terms, symmetric energy term and asymmetric energy term, which respectively correspond to the symmetric and asymmetric components of an object. The asymmetric energy term must be zero, if the studied object is invariant under a reflection about the x-axis. Accordingly, we formulate the problem of detecting reflectional symmetries as a problem of minimizing the asymmetric energy term. From the local minima of the asymmetric energy term, we can detect all the symmetric axes of any object. Since the asymmetric energy term is expressed as a summation of a set of generalized complex (GC) moments computed for an object, the proposed symmetry detection method is robust against both noise and slight deformation. We use the steepest descent technique to calculate the local minima of the asymmetric energy term, whose initialization is calculated from the most dominant GC moment. Experiments on typical logo images and human brain image have shown the effectiveness and the robustness of the proposed method. To our knowledge, this is the first theory on energy functions that describe the symmetric and asymmetric components of a 2D pattern.",
author = "Dinggang Shen and Ip, {Horace H S} and Teoh, {Eam Khwang}",
year = "2001",
month = "4",
day = "1",
doi = "10.1016/S0262-8856(00)00077-9",
language = "English",
volume = "19",
pages = "283--297",
journal = "Image and Vision Computing",
issn = "0262-8856",
publisher = "Elsevier Limited",
number = "5",

}

TY - JOUR

T1 - Energy of asymmetry for accurate detection of global reflection axes

AU - Shen, Dinggang

AU - Ip, Horace H S

AU - Teoh, Eam Khwang

PY - 2001/4/1

Y1 - 2001/4/1

N2 - We define in this paper two energy terms, symmetric energy term and asymmetric energy term, which respectively correspond to the symmetric and asymmetric components of an object. The asymmetric energy term must be zero, if the studied object is invariant under a reflection about the x-axis. Accordingly, we formulate the problem of detecting reflectional symmetries as a problem of minimizing the asymmetric energy term. From the local minima of the asymmetric energy term, we can detect all the symmetric axes of any object. Since the asymmetric energy term is expressed as a summation of a set of generalized complex (GC) moments computed for an object, the proposed symmetry detection method is robust against both noise and slight deformation. We use the steepest descent technique to calculate the local minima of the asymmetric energy term, whose initialization is calculated from the most dominant GC moment. Experiments on typical logo images and human brain image have shown the effectiveness and the robustness of the proposed method. To our knowledge, this is the first theory on energy functions that describe the symmetric and asymmetric components of a 2D pattern.

AB - We define in this paper two energy terms, symmetric energy term and asymmetric energy term, which respectively correspond to the symmetric and asymmetric components of an object. The asymmetric energy term must be zero, if the studied object is invariant under a reflection about the x-axis. Accordingly, we formulate the problem of detecting reflectional symmetries as a problem of minimizing the asymmetric energy term. From the local minima of the asymmetric energy term, we can detect all the symmetric axes of any object. Since the asymmetric energy term is expressed as a summation of a set of generalized complex (GC) moments computed for an object, the proposed symmetry detection method is robust against both noise and slight deformation. We use the steepest descent technique to calculate the local minima of the asymmetric energy term, whose initialization is calculated from the most dominant GC moment. Experiments on typical logo images and human brain image have shown the effectiveness and the robustness of the proposed method. To our knowledge, this is the first theory on energy functions that describe the symmetric and asymmetric components of a 2D pattern.

UR - http://www.scopus.com/inward/record.url?scp=0035314122&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0035314122&partnerID=8YFLogxK

U2 - 10.1016/S0262-8856(00)00077-9

DO - 10.1016/S0262-8856(00)00077-9

M3 - Article

AN - SCOPUS:0035314122

VL - 19

SP - 283

EP - 297

JO - Image and Vision Computing

JF - Image and Vision Computing

SN - 0262-8856

IS - 5

ER -