Globally optimal regions and boundaries as minimum ratio weight cycle

Ian H. Jermyn, Hiroshi Ishikawa

研究成果: Article

130 引用 (Scopus)

抄録

We describe a new form of energy functional for the modelling and identification of regions in images. The energy is defined on the space of boundaries in the image domain and can incorporate very general combinations of modelling information both from the boundary (intensity gradients, etc.) and from the interior of the region (texture, homogeneity, etc.). We describe two polynomial-time digraph algorithms for finding the global minima of this energy. One of the algorithms is completely general, minimizing the functional for any choice of modelling information. It runs in a few seconds on a 256×256 image. The other algorithm applies to a subclass of functionals, but has the advantage of being extremely parallelizable. Neither algorithm requires initialization.

元の言語English
ページ(範囲)1075-1088
ページ数14
ジャーナルIEEE Transactions on Pattern Analysis and Machine Intelligence
23
発行部数10
DOI
出版物ステータスPublished - 2001 10
外部発表Yes

Fingerprint

Cycle
Modeling
Global Minimum
Energy Functional
Energy
Initialization
Digraph
Homogeneity
Texture
Polynomial time
Interior
Textures
Polynomials
Gradient
Form

ASJC Scopus subject areas

  • Control and Systems Engineering
  • Electrical and Electronic Engineering
  • Artificial Intelligence
  • Computer Vision and Pattern Recognition

これを引用

@article{9210ba4a04ae466db24ddd449b02bf72,
title = "Globally optimal regions and boundaries as minimum ratio weight cycle",
abstract = "We describe a new form of energy functional for the modelling and identification of regions in images. The energy is defined on the space of boundaries in the image domain and can incorporate very general combinations of modelling information both from the boundary (intensity gradients, etc.) and from the interior of the region (texture, homogeneity, etc.). We describe two polynomial-time digraph algorithms for finding the global minima of this energy. One of the algorithms is completely general, minimizing the functional for any choice of modelling information. It runs in a few seconds on a 256×256 image. The other algorithm applies to a subclass of functionals, but has the advantage of being extremely parallelizable. Neither algorithm requires initialization.",
keywords = "Active contour, Energy minimization, Global optimum, Ratio, Region identification, Segmentation, Snake",
author = "Jermyn, {Ian H.} and Hiroshi Ishikawa",
year = "2001",
month = "10",
doi = "10.1109/34.954599",
language = "English",
volume = "23",
pages = "1075--1088",
journal = "IEEE Transactions on Pattern Analysis and Machine Intelligence",
issn = "0162-8828",
publisher = "IEEE Computer Society",
number = "10",

}

TY - JOUR

T1 - Globally optimal regions and boundaries as minimum ratio weight cycle

AU - Jermyn, Ian H.

AU - Ishikawa, Hiroshi

PY - 2001/10

Y1 - 2001/10

N2 - We describe a new form of energy functional for the modelling and identification of regions in images. The energy is defined on the space of boundaries in the image domain and can incorporate very general combinations of modelling information both from the boundary (intensity gradients, etc.) and from the interior of the region (texture, homogeneity, etc.). We describe two polynomial-time digraph algorithms for finding the global minima of this energy. One of the algorithms is completely general, minimizing the functional for any choice of modelling information. It runs in a few seconds on a 256×256 image. The other algorithm applies to a subclass of functionals, but has the advantage of being extremely parallelizable. Neither algorithm requires initialization.

AB - We describe a new form of energy functional for the modelling and identification of regions in images. The energy is defined on the space of boundaries in the image domain and can incorporate very general combinations of modelling information both from the boundary (intensity gradients, etc.) and from the interior of the region (texture, homogeneity, etc.). We describe two polynomial-time digraph algorithms for finding the global minima of this energy. One of the algorithms is completely general, minimizing the functional for any choice of modelling information. It runs in a few seconds on a 256×256 image. The other algorithm applies to a subclass of functionals, but has the advantage of being extremely parallelizable. Neither algorithm requires initialization.

KW - Active contour

KW - Energy minimization

KW - Global optimum

KW - Ratio

KW - Region identification

KW - Segmentation

KW - Snake

UR - http://www.scopus.com/inward/record.url?scp=0035481208&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0035481208&partnerID=8YFLogxK

U2 - 10.1109/34.954599

DO - 10.1109/34.954599

M3 - Article

AN - SCOPUS:0035481208

VL - 23

SP - 1075

EP - 1088

JO - IEEE Transactions on Pattern Analysis and Machine Intelligence

JF - IEEE Transactions on Pattern Analysis and Machine Intelligence

SN - 0162-8828

IS - 10

ER -