ScienceDirect® Home Skip Main Navigation Links
You have guest access to ScienceDirect. Find out more.
 
Home
Browse
My Settings
Alerts
Help
 Quick Search
 Search tips (Opens new window)
    Clear all fields    
advertisementadvertisement
Pattern Recognition Letters
Volume 24, Issue 15, November 2003, Pages 2663-2673
 
Font Size: Decrease Font Size  Increase Font Size
 Abstract - selected
Article
Purchase PDF (174 K)

 
 
 
Related Articles in ScienceDirect
View More Related Articles
 
View Record in Scopus
 
doi:10.1016/S0167-8655(03)00109-0    How to Cite or Link Using DOI (Opens New Window)
Copyright © 2003 Elsevier B.V. All rights reserved.

Feature learning with a genetic algorithm for fluorescence fingerprinting of plant species

C. M. CodreaCorresponding Author Contact Information, E-mail The Corresponding Author, a, b, T. Aittokallioa, c, M. Keränend, E. Tyystjärvid and O. S. Nevalainena, b

a Turku Centre for Computer Science, Lemminkäisenkatu 14 A, FIN-20520, Turku, Finland b Department of Information Technology, University of Turku, FIN-20520, Turku, Finland c Department of Mathematics, University of Turku, FIN-20014, Turku, Finland d Department of Biology, Laboratory of Plant Physiology and Molecular Biology, University of Turku, FIN-20014, Turku, Finland

Received 18 November 2002; 
revised 31 March 2003. 
Available online 25 June 2003.

Purchase the full-text article



References and further reading may be available for this article. To view references and further reading you must purchase this article.

Abstract

Proper feature analysis facilitates recognition by focusing the process to those characteristics of observed data that carry the most significant information for the given classification task. In this paper we address the problem of feature selection from a different point of view. Instead of searching for a feature subset out of a large set of predefined candidate features we consider the situation where, given the form of the features and an algorithm for extracting them from the data, the optimizer tunes the feature extraction parameters to improve class separability. This process of feature learning will be solved by the means of a genetic algorithm. The optimized feature set is subsequently used in a neural network classifier. The performance of the feature learning approach is demonstrated with the problem of automatic identification of plant species from their fluorescence induction curves. The general approach should also be useful with other pattern recognition problems where a priori unknown characteristics are extracted from a large feature space.

Author Keywords: Feature analysis; Plant identification; Classification; Fluorescence induction; Genetic algorithm; Neural network

Article Outline

1. Introduction
2. Plant identification from chlorophyll a fluorescence transients
3. Feature learning by a GA
3.1. The stages of the algorithm
3.2. Representation of the solution
3.3. Genetic operators
3.4. Fitness function
3.5. Selection operator
3.6. Initialization and termination
4. Experimental results
4.1. The two class problem
4.1.1. Feature learning using eight line segments
4.1.2. Feature learning using four line segments
4.2. Multiple class problem
5. Conclusions
References







Pattern Recognition Letters
Volume 24, Issue 15, November 2003, Pages 2663-2673
 
Home
Browse
My Settings
Alerts
Help
Elsevier.com (Opens new window)
About ScienceDirect  |  Contact Us  |  Information for Advertisers  |  Terms & Conditions  |  Privacy Policy
Copyright © 2008 Elsevier B.V. All rights reserved. ScienceDirect® is a registered trademark of Elsevier B.V.