Landscape image analysis using fuzzy adaptive resonance theory

Author

Chen, Jia-Lin ; Chang, Jyh- Yeong ; Tsay, Ming-Lay

Author_Institution

Dept. of Electr. & Control Eng., Nat. Chiao Tung Univ., Hsinchu, Taiwan

Volume

3

fYear

1999

fDate

6/21/1905 12:00:00 AM

Firstpage

84

Abstract

Describes an image analysis system using fuzzy methodology. Some landscapes, which are composed of natural things, are chosen as target images. In this system, the fuzzy adaptive resonance theory (fuzzy ART) algorithm is first used to cluster the constituent elements of the landscape images. The means and variance of the clusters are computed and then used to determine the membership functions of fuzzy sets. Based on the derived membership functions, a supervised learning algorithm is used to generate fuzzy rules automatically, and classification will then be facilitated through fuzzy max-min inference. Simulation on real pictures of scenery has shown that the proposed image analyzer is very successful because the result is visually confirmed by human observation

Keywords

adaptive resonance theory; fuzzy logic; fuzzy set theory; image classification; learning (artificial intelligence); pattern clustering; classification; clusters; fuzzy adaptive resonance theory; fuzzy max-min inference; fuzzy rules; human observation; landscape image analysis; membership functions; scenery; supervised learning algorithm; Adaptive systems; Clustering algorithms; Computational modeling; Fuzzy sets; Fuzzy systems; Image analysis; Inference algorithms; Resonance; Subspace constraints; Supervised learning;

fLanguage

English

Publisher

ieee

Conference_Titel

Systems, Man, and Cybernetics, 1999. IEEE SMC '99 Conference Proceedings. 1999 IEEE International Conference on

Conference_Location

Tokyo

ISSN

1062-922X

Print_ISBN

0-7803-5731-0

Type

conf

DOI

10.1109/ICSMC.1999.823159

Filename

823159