High definition feature map for GVF snake by using Harris function

Andrea Kovacs, T. Szirányi

Research output: Chapter in Book/Report/Conference proceedingConference contribution

6 Citations (Scopus)

Abstract

In image segmentation the gradient vector flow snake model is widely used. For concave curvatures snake model has good convergence capabilities, but poor contrast or saddle corner points may result in a loss of contour. We have introduced a new external force component and an optimal initial border, approaching the final boundary as close as possible. We apply keypoints defined by corner functions and their corresponding scale to outline the envelope around the object. The Gradient Vector Flow (GVF) field is generated by the eigenvalues of Harris matrix and/or the scale of the feature point. The GVF field is featured by new functions characterizing the edginess and cornerness in one function. We have shown that the max(0,log[max(λ 1, λ 2)]) function fulfills the requirements for any active contour definitions in case of difficult shapes and background conditions. This new GVF field has several advantages: smooth transitions are robustly taken into account, while sharp corners and contour scragginess can be perfectly detected.

Original languageEnglish
Title of host publicationLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Pages163-172
Number of pages10
Volume6474 LNCS
EditionPART 1
DOIs
Publication statusPublished - 2010
Event12th International Conference on Advanced Concepts for Intelligent Vision Systems, ACIVS 2010 - Sydney, NSW, Australia
Duration: Dec 13 2010Dec 16 2010

Publication series

NameLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
NumberPART 1
Volume6474 LNCS
ISSN (Print)03029743
ISSN (Electronic)16113349

Other

Other12th International Conference on Advanced Concepts for Intelligent Vision Systems, ACIVS 2010
CountryAustralia
CitySydney, NSW
Period12/13/1012/16/10

Fingerprint

Gradient Vector Flow
Snakes
Flow Field
Vector Field
Flow fields
Active Contours
Feature Point
Saddle
Image segmentation
Image Segmentation
Envelope
Curvature
Eigenvalue
Requirements
Model

Keywords

  • Active-contour
  • Corner detection
  • Harris function
  • Shape Analysis

ASJC Scopus subject areas

  • Computer Science(all)
  • Theoretical Computer Science

Cite this

Kovacs, A., & Szirányi, T. (2010). High definition feature map for GVF snake by using Harris function. In Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) (PART 1 ed., Vol. 6474 LNCS, pp. 163-172). (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 6474 LNCS, No. PART 1). https://doi.org/10.1007/978-3-642-17688-3_17

High definition feature map for GVF snake by using Harris function. / Kovacs, Andrea; Szirányi, T.

Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics). Vol. 6474 LNCS PART 1. ed. 2010. p. 163-172 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 6474 LNCS, No. PART 1).

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Kovacs, A & Szirányi, T 2010, High definition feature map for GVF snake by using Harris function. in Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics). PART 1 edn, vol. 6474 LNCS, Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), no. PART 1, vol. 6474 LNCS, pp. 163-172, 12th International Conference on Advanced Concepts for Intelligent Vision Systems, ACIVS 2010, Sydney, NSW, Australia, 12/13/10. https://doi.org/10.1007/978-3-642-17688-3_17
Kovacs A, Szirányi T. High definition feature map for GVF snake by using Harris function. In Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics). PART 1 ed. Vol. 6474 LNCS. 2010. p. 163-172. (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); PART 1). https://doi.org/10.1007/978-3-642-17688-3_17
Kovacs, Andrea ; Szirányi, T. / High definition feature map for GVF snake by using Harris function. Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics). Vol. 6474 LNCS PART 1. ed. 2010. pp. 163-172 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); PART 1).
@inproceedings{58410a7d5e5848bfa7bb0ff9661fcff4,
title = "High definition feature map for GVF snake by using Harris function",
abstract = "In image segmentation the gradient vector flow snake model is widely used. For concave curvatures snake model has good convergence capabilities, but poor contrast or saddle corner points may result in a loss of contour. We have introduced a new external force component and an optimal initial border, approaching the final boundary as close as possible. We apply keypoints defined by corner functions and their corresponding scale to outline the envelope around the object. The Gradient Vector Flow (GVF) field is generated by the eigenvalues of Harris matrix and/or the scale of the feature point. The GVF field is featured by new functions characterizing the edginess and cornerness in one function. We have shown that the max(0,log[max(λ 1, λ 2)]) function fulfills the requirements for any active contour definitions in case of difficult shapes and background conditions. This new GVF field has several advantages: smooth transitions are robustly taken into account, while sharp corners and contour scragginess can be perfectly detected.",
keywords = "Active-contour, Corner detection, Harris function, Shape Analysis",
author = "Andrea Kovacs and T. Szir{\'a}nyi",
year = "2010",
doi = "10.1007/978-3-642-17688-3_17",
language = "English",
isbn = "3642176879",
volume = "6474 LNCS",
series = "Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)",
number = "PART 1",
pages = "163--172",
booktitle = "Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)",
edition = "PART 1",

}

TY - GEN

T1 - High definition feature map for GVF snake by using Harris function

AU - Kovacs, Andrea

AU - Szirányi, T.

PY - 2010

Y1 - 2010

N2 - In image segmentation the gradient vector flow snake model is widely used. For concave curvatures snake model has good convergence capabilities, but poor contrast or saddle corner points may result in a loss of contour. We have introduced a new external force component and an optimal initial border, approaching the final boundary as close as possible. We apply keypoints defined by corner functions and their corresponding scale to outline the envelope around the object. The Gradient Vector Flow (GVF) field is generated by the eigenvalues of Harris matrix and/or the scale of the feature point. The GVF field is featured by new functions characterizing the edginess and cornerness in one function. We have shown that the max(0,log[max(λ 1, λ 2)]) function fulfills the requirements for any active contour definitions in case of difficult shapes and background conditions. This new GVF field has several advantages: smooth transitions are robustly taken into account, while sharp corners and contour scragginess can be perfectly detected.

AB - In image segmentation the gradient vector flow snake model is widely used. For concave curvatures snake model has good convergence capabilities, but poor contrast or saddle corner points may result in a loss of contour. We have introduced a new external force component and an optimal initial border, approaching the final boundary as close as possible. We apply keypoints defined by corner functions and their corresponding scale to outline the envelope around the object. The Gradient Vector Flow (GVF) field is generated by the eigenvalues of Harris matrix and/or the scale of the feature point. The GVF field is featured by new functions characterizing the edginess and cornerness in one function. We have shown that the max(0,log[max(λ 1, λ 2)]) function fulfills the requirements for any active contour definitions in case of difficult shapes and background conditions. This new GVF field has several advantages: smooth transitions are robustly taken into account, while sharp corners and contour scragginess can be perfectly detected.

KW - Active-contour

KW - Corner detection

KW - Harris function

KW - Shape Analysis

UR - http://www.scopus.com/inward/record.url?scp=78650855116&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=78650855116&partnerID=8YFLogxK

U2 - 10.1007/978-3-642-17688-3_17

DO - 10.1007/978-3-642-17688-3_17

M3 - Conference contribution

AN - SCOPUS:78650855116

SN - 3642176879

SN - 9783642176876

VL - 6474 LNCS

T3 - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)

SP - 163

EP - 172

BT - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)

ER -