TRAINING-BASED OPTIMIZATION OF SOFT MORPHOLOGICAL FILTERS

Authors
KOIVISTO P HUTTUNEN H KUOSMANEN P
Citation
P. Koivisto et al., TRAINING-BASED OPTIMIZATION OF SOFT MORPHOLOGICAL FILTERS, Journal of electronic imaging, 5(3), 1996, pp. 300-322
Citations number
18
Categorie Soggetti
Engineering, Eletrical & Electronic",Optics,"Photographic Tecnology
Journal title
Journal of electronic imaging → ACNP
ISSN journal
10179909
Volume
5
Issue
3
Year of publication
1996
Pages
300 - 322
Database
ISI
SICI code
1017-9909(1996)5:3<300:TOOSMF>2.0.ZU;2-T
Abstract
Soft morphological filters farm a large class of nonlinear filters wit h many desirable properties. However, few design methods exist for the se filters. This paper demonstrates how optimization schemes. simulate d annealing and genetic algorithms, can be employed in the search for soft morphological filters having optimal performance in a given signa l processing task. Furthermore, the properties of the achieved optimal soft morphological filters in different situations are analyzed. (C) 1996 SPIE and IS&T.