Natural image profiles are most likely to be step edges
Research output: Contribution to journal › Journal article › Research › peer-review
Standard
Natural image profiles are most likely to be step edges. / Griffin, Lewis D.; Lillholm, Martin; Nielsen, Mads.
In: Vision Research, Vol. 44, No. 4, 2004, p. 407-421.Research output: Contribution to journal › Journal article › Research › peer-review
Harvard
APA
Vancouver
Author
Bibtex
}
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