An adaptive approach to scale selection for line and edge detection

M. Concetta Morrone, Anacleto Navangione, David Burr

Research output: Contribution to journalArticle

Abstract

One of the standard problems of edge- and line-detecting algorithms is to determine the most appropriate size of the convolution-operator for the particular task, maximising the conflicting goals of resolution and sensitivity. Here we suggest a novel approach to scale selection, where the scale size varies dynamically with the convolution output: the stronger the output, the smaller the spatial scale. This principle has been applied to two types of feature-detection algorithms, and shown to perform well for both one- and two-dimensional images.

Original languageEnglish
Pages (from-to)667-677
Number of pages11
JournalPattern Recognition Letters
Volume16
Issue number7
DOIs
Publication statusPublished - 1995

Fingerprint

Edge detection
Convolution
Mathematical operators

Keywords

  • Adaptive algorithms
  • Edge detection
  • Feature detection
  • Local energy function
  • Quadrature filters
  • Visual receptive field size

ASJC Scopus subject areas

  • Artificial Intelligence
  • Computer Vision and Pattern Recognition
  • Signal Processing
  • Software
  • Electrical and Electronic Engineering

Cite this

An adaptive approach to scale selection for line and edge detection. / Morrone, M. Concetta; Navangione, Anacleto; Burr, David.

In: Pattern Recognition Letters, Vol. 16, No. 7, 1995, p. 667-677.

Research output: Contribution to journalArticle

Morrone, M. Concetta ; Navangione, Anacleto ; Burr, David. / An adaptive approach to scale selection for line and edge detection. In: Pattern Recognition Letters. 1995 ; Vol. 16, No. 7. pp. 667-677.
@article{0766b799d0b54ac080f7b40b8e7c97f4,
title = "An adaptive approach to scale selection for line and edge detection",
abstract = "One of the standard problems of edge- and line-detecting algorithms is to determine the most appropriate size of the convolution-operator for the particular task, maximising the conflicting goals of resolution and sensitivity. Here we suggest a novel approach to scale selection, where the scale size varies dynamically with the convolution output: the stronger the output, the smaller the spatial scale. This principle has been applied to two types of feature-detection algorithms, and shown to perform well for both one- and two-dimensional images.",
keywords = "Adaptive algorithms, Edge detection, Feature detection, Local energy function, Quadrature filters, Visual receptive field size",
author = "Morrone, {M. Concetta} and Anacleto Navangione and David Burr",
year = "1995",
doi = "10.1016/0167-8655(95)00017-B",
language = "English",
volume = "16",
pages = "667--677",
journal = "Pattern Recognition Letters",
issn = "0167-8655",
publisher = "Elsevier",
number = "7",

}

TY - JOUR

T1 - An adaptive approach to scale selection for line and edge detection

AU - Morrone, M. Concetta

AU - Navangione, Anacleto

AU - Burr, David

PY - 1995

Y1 - 1995

N2 - One of the standard problems of edge- and line-detecting algorithms is to determine the most appropriate size of the convolution-operator for the particular task, maximising the conflicting goals of resolution and sensitivity. Here we suggest a novel approach to scale selection, where the scale size varies dynamically with the convolution output: the stronger the output, the smaller the spatial scale. This principle has been applied to two types of feature-detection algorithms, and shown to perform well for both one- and two-dimensional images.

AB - One of the standard problems of edge- and line-detecting algorithms is to determine the most appropriate size of the convolution-operator for the particular task, maximising the conflicting goals of resolution and sensitivity. Here we suggest a novel approach to scale selection, where the scale size varies dynamically with the convolution output: the stronger the output, the smaller the spatial scale. This principle has been applied to two types of feature-detection algorithms, and shown to perform well for both one- and two-dimensional images.

KW - Adaptive algorithms

KW - Edge detection

KW - Feature detection

KW - Local energy function

KW - Quadrature filters

KW - Visual receptive field size

UR - http://www.scopus.com/inward/record.url?scp=0029343955&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0029343955&partnerID=8YFLogxK

U2 - 10.1016/0167-8655(95)00017-B

DO - 10.1016/0167-8655(95)00017-B

M3 - Article

AN - SCOPUS:0029343955

VL - 16

SP - 667

EP - 677

JO - Pattern Recognition Letters

JF - Pattern Recognition Letters

SN - 0167-8655

IS - 7

ER -