Natural image profiles are most likely to be step edges

Publikation: Bidrag til tidsskriftTidsskriftartikelfagfællebedømt

Standard

Natural image profiles are most likely to be step edges. / Griffin, Lewis D.; Lillholm, Martin; Nielsen, Mads.

I: Vision Research, Bind 44, Nr. 4, 2004, s. 407-421.

Publikation: Bidrag til tidsskriftTidsskriftartikelfagfællebedømt

Harvard

Griffin, LD, Lillholm, M & Nielsen, M 2004, 'Natural image profiles are most likely to be step edges', Vision Research, bind 44, nr. 4, s. 407-421. https://doi.org/10.1016/j.visres.2003.09.025

APA

Griffin, L. D., Lillholm, M., & Nielsen, M. (2004). Natural image profiles are most likely to be step edges. Vision Research, 44(4), 407-421. https://doi.org/10.1016/j.visres.2003.09.025

Vancouver

Griffin LD, Lillholm M, Nielsen M. Natural image profiles are most likely to be step edges. Vision Research. 2004;44(4):407-421. https://doi.org/10.1016/j.visres.2003.09.025

Author

Griffin, Lewis D. ; Lillholm, Martin ; Nielsen, Mads. / Natural image profiles are most likely to be step edges. I: Vision Research. 2004 ; Bind 44, Nr. 4. s. 407-421.

Bibtex

@article{1fb196e06dc511dd8d9f000ea68e967b,
title = "Natural image profiles are most likely to be step edges",
abstract = "We introduce Geometric Texton Theory (GTT), a theory of categorical visual feature classification that arises through consideration of the metamerism that affects families of co-localised linear receptive-field operators. A refinement of GTT that uses maximum likelihood (ML) to resolve this metamerism is presented. We describe a method for discovering the ML element of a metamery class by analysing a database of natural images. We apply the method to the simplest case––the ML element of a canonical metamery class defined by co-registering the location and orientation of profiles from images, and affinely scaling their intensities so that they have identical responses to 1-D, zeroth- and first-order, derivative of Gaussian operators. We find that a step edge is the ML profile. This result is consistent with our proposed theory of feature classification.",
author = "Griffin, {Lewis D.} and Martin Lillholm and Mads Nielsen",
year = "2004",
doi = "10.1016/j.visres.2003.09.025",
language = "English",
volume = "44",
pages = "407--421",
journal = "Vision Research",
issn = "0042-6989",
publisher = "Pergamon Press",
number = "4",

}

RIS

TY - JOUR

T1 - Natural image profiles are most likely to be step edges

AU - Griffin, Lewis D.

AU - Lillholm, Martin

AU - Nielsen, Mads

PY - 2004

Y1 - 2004

N2 - We introduce Geometric Texton Theory (GTT), a theory of categorical visual feature classification that arises through consideration of the metamerism that affects families of co-localised linear receptive-field operators. A refinement of GTT that uses maximum likelihood (ML) to resolve this metamerism is presented. We describe a method for discovering the ML element of a metamery class by analysing a database of natural images. We apply the method to the simplest case––the ML element of a canonical metamery class defined by co-registering the location and orientation of profiles from images, and affinely scaling their intensities so that they have identical responses to 1-D, zeroth- and first-order, derivative of Gaussian operators. We find that a step edge is the ML profile. This result is consistent with our proposed theory of feature classification.

AB - We introduce Geometric Texton Theory (GTT), a theory of categorical visual feature classification that arises through consideration of the metamerism that affects families of co-localised linear receptive-field operators. A refinement of GTT that uses maximum likelihood (ML) to resolve this metamerism is presented. We describe a method for discovering the ML element of a metamery class by analysing a database of natural images. We apply the method to the simplest case––the ML element of a canonical metamery class defined by co-registering the location and orientation of profiles from images, and affinely scaling their intensities so that they have identical responses to 1-D, zeroth- and first-order, derivative of Gaussian operators. We find that a step edge is the ML profile. This result is consistent with our proposed theory of feature classification.

U2 - 10.1016/j.visres.2003.09.025

DO - 10.1016/j.visres.2003.09.025

M3 - Journal article

VL - 44

SP - 407

EP - 421

JO - Vision Research

JF - Vision Research

SN - 0042-6989

IS - 4

ER -

ID: 5580602