Using restricted loops in genetic programming for image classification

Wijesinghe, G and Ciesielski, V 2007, 'Using restricted loops in genetic programming for image classification', in L. Wang (ed.) IEEE Congress on Evolutionary Computation, Singapore, 25-28 September 2007, pp. 4569-4576.


Document type: Conference Paper
Collection: Conference Papers

Attached Files
Name Description MIMEType Size
n2006006527.pdf Published version application/pdf 276.83KB
Title Using restricted loops in genetic programming for image classification
Author(s) Wijesinghe, G
Ciesielski, V
Year 2007
Conference name IEEE Congress on Evolutionary Computation
Conference location Singapore
Conference dates 25-28 September 2007
Proceedings title IEEE Congress on Evolutionary Computation
Editor(s) L. Wang
Publisher IEEE
Place of publication Piscataway, USA
Start page 4569
End page 4576
Total pages 8
Abstract Loops are rarely used in genetic programming due to issues such as detecting infinite loops and invalid programs. In this paper we present a restricted form of loops that is specifically designed to be evolved in image classifiers. Particularly, we evolve classifiers that use these loops to perform calculations on image regions chosen by the loops. We have compared this method to another classification method that only uses individual pixels in its calculations. These two methods are tested on two synthesised and one non-synthesised greyscale image classification problems of varying difficulty. The most difficult problem requires determining heads or tails of 320 x 320 pixel images of a US one cent coin at any angle. On this problem, the accuracy of the loops approach was 97.80% in contrast to the no-loop method accuracy of 79.46%. Use of loops also reduces overfitting of training data. We also found that loop methods overfit less when only a few training examples are available.
Subjects Neural, Evolutionary and Fuzzy Computation
DOI - identifier 10.1109/CEC.2007.4425070
Copyright notice © 2007 IEEE
ISBN 978-1-4244-1339-3
Versions
Version Filter Type
Citation counts: Scopus Citation Count Cited 7 times in Scopus Article | Citations
Altmetric details:
Access Statistics: 173 Abstract Views, 421 File Downloads  -  Detailed Statistics
Created: Wed, 08 Apr 2009, 09:42:32 EST by Catalyst Administrator
© 2014 RMIT Research Repository • Powered by Fez SoftwareContact us