Learning fuzzy modelling through genetic algorithm for object recognition

Author

Soodamani, R. ; Liu, Z.Q.

Author_Institution

Dept. of Comput. Sci., Melbourne Univ., Australia

fYear

1998

fDate

4-9 May 1998

Firstpage

656

Lastpage

659

Abstract

The paper proposes a genetic-algorithm-based learning strategy that models membership functions of the fuzzy attributes of surfaces in a model based machine vision system. The objective function aims at enhancing recognition performance in terms of maximizing the degree of discrimination among classes. As a result, the accuracy of recognizing known instances of objects and generalisation capability by recognizing unknown instances of known objects are greatly improved. The performance enhancement of a model based object recognition system consisting of a set of synthetic range images is established by incorporating a dynamic off-line learning mechanism using a genetic algorithm in the feedback path of the system

Keywords

computer vision; feedback; fuzzy systems; generalisation (artificial intelligence); genetic algorithms; learning (artificial intelligence); object recognition; class discrimination; dynamic off-line learning mechanism; feedback path; fuzzy attributes; fuzzy modelling learning; generalisation capability; genetic algorithm; known objects; membership functions; model based machine vision system; object recognition; objective function; surfaces; synthetic range images; unknown instance recognition; Computer vision; Feedback; Fuzzy sets; Fuzzy systems; Genetic algorithms; Image segmentation; Machine intelligence; Object recognition; Shape; System testing;

fLanguage

English

Publisher

ieee

Conference_Titel

Evolutionary Computation Proceedings, 1998. IEEE World Congress on Computational Intelligence., The 1998 IEEE International Conference on

Conference_Location

Anchorage, AK

Print_ISBN

0-7803-4869-9

Type

conf

DOI

10.1109/ICEC.1998.700117

Filename

700117