Genetic algorithms for automatic feature selection in a textureclassification system

Stolpmann, Alexander and Dooley, Laurence S. (1998). Genetic algorithms for automatic feature selection in a textureclassification system. In: 4th IEEE International Conference on Signal Processing (ICSP'98), 12th-16th Oct 1998, Beijing, China.

Full text available as:

Preview

PDF (Not Set) - Requires a PDF viewer such as GSview, Xpdf or Adobe Acrobat Reader
Download (427Kb)

DOI (Digital Object Identifier) Link:	http://dx.doi.org/doi:10.1109/ICOSP.1998.770840
Google Scholar:	Look up in Google Scholar

Abstract

This paper describes the use of genetic algorithms as feature selectors in a texture classification system. This is part of a system developed within a research project concerning the classification of genuine texture. An attempt is made to underline why an automatic feature selector is a useful part of the texture classification system. Furthermore a way of including the genetic algorithms into the system and the necessary feedback structure is explained

Item Type:	Conference Item
ISBN:	0-7803-4325-5, 978-0-7803-4325-2
Keywords:	feature extraction; feedback; genetic algorithms; image classification; image texture; automatic feature selection; feature selectors; feedback structure; genetic algorithms; texture classification system
Academic Unit/Department:	Mathematics, Computing and Technology > Communication and Systems
Interdisciplinary Research Centre:	Centre for Research in Computing (CRC)
Item ID:	16640
Depositing User:	Laurence Dooley
Date Deposited:	02 Jun 2009 14:04
Last Modified:	05 Dec 2010 06:47
URI:	http://oro.open.ac.uk/id/eprint/16640

Actions (login may be required)

	View Item
	Public: Report issue / request change

Policies | Disclaimer