Recognition model with narrow and broad extension fields

Authors
Citation
P. Kalocsai, Recognition model with narrow and broad extension fields, INF SCI, 126(1-4), 2000, pp. 41-56
Citations number
18
Categorie Soggetti
Information Tecnology & Communication Systems
Journal title
INFORMATION SCIENCES
ISSN journal
00200255 → ACNP
Volume
126
Issue
1-4
Year of publication
2000
Pages
41 - 56
Database
ISI
SICI code
0020-0255(200007)126:1-4<41:RMWNAB>2.0.ZU;2-N
Abstract
A recognition model which defines a measure of shape similarity on the dire ct output of multiscale and multiorientation Gabor filters does not manifes t qualitative aspects of human object recognition of contour-deleted images in that: (a) it recognizes recoverable and nonrecoverable contour-deleted images equally well whereas humans recognize recoverable images much better , (b) it distinguishes complementary feature-deleted images whereas humans do not. Adding some of the known connectivity pattern of the primary visual cortex to the model in the form of extension fields (connections between c ollinear and curvilinear units) among filters increased the overall recogni tion performance of the model and: (a) boosted the recognition rate of the recoverable images far more than the nonrecoverable ones, and (b) increased the similarity of complementary feature-deleted images, but not part-delet ed ones, and thus attained a closer correspondence to human psychophysical results. Interestingly, performance was approximately equivalent for narrow (+/-15 degrees) and broad (+/-90 degrees) extension fields, (C) 2000 Publi shed by Elsevier Science Inc. Al rights reserved.