Mach edges: local features predicted by 3rd derivative spatial filtering

Stuart A. Wallis, Mark A. Georgeson

Research output: Contribution to journalArticle

Abstract

Edges are key points of information in visual scenes. One important class of models supposes that edges correspond to the steepest parts of the luminance profile, implying that they can be found as peaks and troughs in the response of a gradient (1st derivative) filter, or as zero-crossings in the 2nd derivative (ZCs). We tested those ideas using a stimulus that has no local peaks of gradient and no ZCs, at any scale. The stimulus profile is analogous to the Mach ramp, but it is the luminance gradient (not the absolute luminance) that increases as a linear ramp between two plateaux; the luminance profile is a blurred triangle-wave. For all image-blurs tested, observers marked edges at or close to the corner points in the gradient profile, even though these were not gradient maxima. These Mach edges correspond to peaks and troughs in the 3rd derivative. Thus Mach edges are inconsistent with many standard edge-detection schemes, but are nicely predicted by a recent model that finds edge points with a 2-stage sequence of 1st then 2nd derivative operators, each followed by a half-wave rectifier.
LanguageEnglish
Pages1886-1893
Number of pages8
JournalVision Research
Volume49
Issue number14
DOIs
Publication statusPublished - Jul 2009

Fingerprint

Architectural Accessibility

Bibliographical note

NOTICE: this is the author’s version of a work that was accepted for publication in Vision Research. Changes resulting from the publishing process, such as peer review, editing, corrections, structural formatting, and other quality control mechanisms may not be reflected in this document. Changes may have been made to this work since it was submitted for publication. A definitive version was subsequently published in Wallis, Stuart A. and Georgeson, Mark A. (2009). Mach edges: local features predicted by 3rd derivative spatial filtering. Vision Research, 49 (14), pp. 1886-1893. DOI 10.1016/j.visres.2009.04.026

Keywords

  • human vision
  • psychophysics
  • edge localisation
  • spatial derivatives
  • Mach bands

Cite this

Wallis, Stuart A. ; Georgeson, Mark A. / Mach edges: local features predicted by 3rd derivative spatial filtering. In: Vision Research. 2009 ; Vol. 49, No. 14. pp. 1886-1893.
@article{134d3ccad0274e01a5156a91502f2952,
title = "Mach edges: local features predicted by 3rd derivative spatial filtering",
abstract = "Edges are key points of information in visual scenes. One important class of models supposes that edges correspond to the steepest parts of the luminance profile, implying that they can be found as peaks and troughs in the response of a gradient (1st derivative) filter, or as zero-crossings in the 2nd derivative (ZCs). We tested those ideas using a stimulus that has no local peaks of gradient and no ZCs, at any scale. The stimulus profile is analogous to the Mach ramp, but it is the luminance gradient (not the absolute luminance) that increases as a linear ramp between two plateaux; the luminance profile is a blurred triangle-wave. For all image-blurs tested, observers marked edges at or close to the corner points in the gradient profile, even though these were not gradient maxima. These Mach edges correspond to peaks and troughs in the 3rd derivative. Thus Mach edges are inconsistent with many standard edge-detection schemes, but are nicely predicted by a recent model that finds edge points with a 2-stage sequence of 1st then 2nd derivative operators, each followed by a half-wave rectifier.",
keywords = "human vision, psychophysics, edge localisation, spatial derivatives, Mach bands",
author = "Wallis, {Stuart A.} and Georgeson, {Mark A.}",
note = "NOTICE: this is the author’s version of a work that was accepted for publication in Vision Research. Changes resulting from the publishing process, such as peer review, editing, corrections, structural formatting, and other quality control mechanisms may not be reflected in this document. Changes may have been made to this work since it was submitted for publication. A definitive version was subsequently published in Wallis, Stuart A. and Georgeson, Mark A. (2009). Mach edges: local features predicted by 3rd derivative spatial filtering. Vision Research, 49 (14), pp. 1886-1893. DOI 10.1016/j.visres.2009.04.026",
year = "2009",
month = "7",
doi = "10.1016/j.visres.2009.04.026",
language = "English",
volume = "49",
pages = "1886--1893",
journal = "Vision Research",
issn = "0042-6989",
publisher = "Elsevier",
number = "14",

}

Mach edges: local features predicted by 3rd derivative spatial filtering. / Wallis, Stuart A.; Georgeson, Mark A.

In: Vision Research, Vol. 49, No. 14, 07.2009, p. 1886-1893.

Research output: Contribution to journalArticle

TY - JOUR

T1 - Mach edges: local features predicted by 3rd derivative spatial filtering

AU - Wallis, Stuart A.

AU - Georgeson, Mark A.

N1 - NOTICE: this is the author’s version of a work that was accepted for publication in Vision Research. Changes resulting from the publishing process, such as peer review, editing, corrections, structural formatting, and other quality control mechanisms may not be reflected in this document. Changes may have been made to this work since it was submitted for publication. A definitive version was subsequently published in Wallis, Stuart A. and Georgeson, Mark A. (2009). Mach edges: local features predicted by 3rd derivative spatial filtering. Vision Research, 49 (14), pp. 1886-1893. DOI 10.1016/j.visres.2009.04.026

PY - 2009/7

Y1 - 2009/7

N2 - Edges are key points of information in visual scenes. One important class of models supposes that edges correspond to the steepest parts of the luminance profile, implying that they can be found as peaks and troughs in the response of a gradient (1st derivative) filter, or as zero-crossings in the 2nd derivative (ZCs). We tested those ideas using a stimulus that has no local peaks of gradient and no ZCs, at any scale. The stimulus profile is analogous to the Mach ramp, but it is the luminance gradient (not the absolute luminance) that increases as a linear ramp between two plateaux; the luminance profile is a blurred triangle-wave. For all image-blurs tested, observers marked edges at or close to the corner points in the gradient profile, even though these were not gradient maxima. These Mach edges correspond to peaks and troughs in the 3rd derivative. Thus Mach edges are inconsistent with many standard edge-detection schemes, but are nicely predicted by a recent model that finds edge points with a 2-stage sequence of 1st then 2nd derivative operators, each followed by a half-wave rectifier.

AB - Edges are key points of information in visual scenes. One important class of models supposes that edges correspond to the steepest parts of the luminance profile, implying that they can be found as peaks and troughs in the response of a gradient (1st derivative) filter, or as zero-crossings in the 2nd derivative (ZCs). We tested those ideas using a stimulus that has no local peaks of gradient and no ZCs, at any scale. The stimulus profile is analogous to the Mach ramp, but it is the luminance gradient (not the absolute luminance) that increases as a linear ramp between two plateaux; the luminance profile is a blurred triangle-wave. For all image-blurs tested, observers marked edges at or close to the corner points in the gradient profile, even though these were not gradient maxima. These Mach edges correspond to peaks and troughs in the 3rd derivative. Thus Mach edges are inconsistent with many standard edge-detection schemes, but are nicely predicted by a recent model that finds edge points with a 2-stage sequence of 1st then 2nd derivative operators, each followed by a half-wave rectifier.

KW - human vision

KW - psychophysics

KW - edge localisation

KW - spatial derivatives

KW - Mach bands

UR - http://www.scopus.com/inward/record.url?scp=67349176017&partnerID=8YFLogxK

U2 - 10.1016/j.visres.2009.04.026

DO - 10.1016/j.visres.2009.04.026

M3 - Article

VL - 49

SP - 1886

EP - 1893

JO - Vision Research

T2 - Vision Research

JF - Vision Research

SN - 0042-6989

IS - 14

ER -