Using Restricted Loops in Genetic Programming for Image Classification

Created by W.Langdon from gp-bibliography.bib Revision:1.4221

  author =       "Gayan Wijesinghe and Vic Ciesielski",
  title =        "Using Restricted Loops in Genetic Programming for
                 Image Classification",
  booktitle =    "2007 IEEE Congress on Evolutionary Computation",
  year =         "2007",
  editor =       "Dipti Srinivasan and Lipo Wang",
  pages =        "4569--4576",
  address =      "Singapore",
  month =        "25-28 " # sep,
  organization = "IEEE Computational Intelligence Society",
  publisher =    "IEEE Press",
  keywords =     "genetic algorithms, genetic programming, STGP,
                 greyscale image classification, infinite loops, invalid
                 programs, no-loop method accuracy, restricted loops,
                 image classification",
  ISBN =         "1-4244-1340-0",
  file =         "1664.pdf",
  DOI =          "doi:10.1109/CEC.2007.4425070",
  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 grey scale image classification
                 problems of varying difficulty. The most difficult
                 problem requires determining heads or tails of 320 by
                 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 over fitting of
                 training data. We also found that loop methods overfit
                 less when only a few training examples are available.",
  notes =        "CEC 2007 - A joint meeting of the IEEE, the EPS, and
                 the IET.

                 IEEE Catalog Number: 07TH8963C",

Genetic Programming entries for Gayan Wijesinghe Victor Ciesielski