Generalized gradient vector flow external forces for active contours

Chenyang Xu, Jerry Ladd Prince

Research output: Contribution to journalArticle

Abstract

Active contours, or snakes, are used extensively in computer vision and image processing applications, particularly to locate object boundaries. A new type of external force for active contours, called gradient vector flow (GVF) was introduced recently to address problems associated with initialization and poor convergence to boundary concavities. GVF is computed as a diffusion of the gradient vectors of a gray-level or binary edge map derived from the image. In this paper, we generalize the GVF formulation to include two spatially varying weighting functions. This improves active contour convergence to long, thin boundary indentations, while maintaining other desirable properties of GVF, such as an extended capture range. The original GVF is a special case of this new generalized GVF (GGVF) model. An error analysis for active contour results on simulated test images is also presented.

Original languageEnglish (US)
Pages (from-to)131-139
Number of pages9
JournalSignal Processing
Volume71
Issue number2
StatePublished - Dec 15 1998

Fingerprint

Indentation
Error analysis
Computer vision
Image processing

Keywords

  • Active contour models
  • Deformable models
  • Edge detection
  • Gradient vector flow
  • Image segmentation
  • Shape representation and recovery

ASJC Scopus subject areas

  • Signal Processing
  • Electrical and Electronic Engineering

Cite this

Generalized gradient vector flow external forces for active contours. / Xu, Chenyang; Prince, Jerry Ladd.

In: Signal Processing, Vol. 71, No. 2, 15.12.1998, p. 131-139.

Research output: Contribution to journalArticle

@article{8c5cfae6f03e4711992875d74547d735,
title = "Generalized gradient vector flow external forces for active contours",
abstract = "Active contours, or snakes, are used extensively in computer vision and image processing applications, particularly to locate object boundaries. A new type of external force for active contours, called gradient vector flow (GVF) was introduced recently to address problems associated with initialization and poor convergence to boundary concavities. GVF is computed as a diffusion of the gradient vectors of a gray-level or binary edge map derived from the image. In this paper, we generalize the GVF formulation to include two spatially varying weighting functions. This improves active contour convergence to long, thin boundary indentations, while maintaining other desirable properties of GVF, such as an extended capture range. The original GVF is a special case of this new generalized GVF (GGVF) model. An error analysis for active contour results on simulated test images is also presented.",
keywords = "Active contour models, Deformable models, Edge detection, Gradient vector flow, Image segmentation, Shape representation and recovery",
author = "Chenyang Xu and Prince, {Jerry Ladd}",
year = "1998",
month = "12",
day = "15",
language = "English (US)",
volume = "71",
pages = "131--139",
journal = "Signal Processing",
issn = "0165-1684",
publisher = "Elsevier",
number = "2",

}

TY - JOUR

T1 - Generalized gradient vector flow external forces for active contours

AU - Xu, Chenyang

AU - Prince, Jerry Ladd

PY - 1998/12/15

Y1 - 1998/12/15

N2 - Active contours, or snakes, are used extensively in computer vision and image processing applications, particularly to locate object boundaries. A new type of external force for active contours, called gradient vector flow (GVF) was introduced recently to address problems associated with initialization and poor convergence to boundary concavities. GVF is computed as a diffusion of the gradient vectors of a gray-level or binary edge map derived from the image. In this paper, we generalize the GVF formulation to include two spatially varying weighting functions. This improves active contour convergence to long, thin boundary indentations, while maintaining other desirable properties of GVF, such as an extended capture range. The original GVF is a special case of this new generalized GVF (GGVF) model. An error analysis for active contour results on simulated test images is also presented.

AB - Active contours, or snakes, are used extensively in computer vision and image processing applications, particularly to locate object boundaries. A new type of external force for active contours, called gradient vector flow (GVF) was introduced recently to address problems associated with initialization and poor convergence to boundary concavities. GVF is computed as a diffusion of the gradient vectors of a gray-level or binary edge map derived from the image. In this paper, we generalize the GVF formulation to include two spatially varying weighting functions. This improves active contour convergence to long, thin boundary indentations, while maintaining other desirable properties of GVF, such as an extended capture range. The original GVF is a special case of this new generalized GVF (GGVF) model. An error analysis for active contour results on simulated test images is also presented.

KW - Active contour models

KW - Deformable models

KW - Edge detection

KW - Gradient vector flow

KW - Image segmentation

KW - Shape representation and recovery

UR - http://www.scopus.com/inward/record.url?scp=0032291262&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0032291262&partnerID=8YFLogxK

M3 - Article

VL - 71

SP - 131

EP - 139

JO - Signal Processing

JF - Signal Processing

SN - 0165-1684

IS - 2

ER -