Detail publikace

Acceleration of Evolutionary Image Filter Design Using Coevolution in Cartesian GP

Originální název

Acceleration of Evolutionary Image Filter Design Using Coevolution in Cartesian GP

Anglický název

Acceleration of Evolutionary Image Filter Design Using Coevolution in Cartesian GP

Jazyk

en

Originální abstrakt

The aim of this paper is to accelerate the task of evolutionary image filter design using coevolution of candidate filters and training vectors subsets. Two coevolutionary methods are implemented and compared for this task in the framework of Cartesian Genetic Programming (CGP). Experimental results show that only 15-20 % of original test vectors are needed to find an image filter which provides the same quality of filtering as the best filter evolved using the standard CGP which utilizes the whole training set. Moreover, the median time of evolution was reduced 2.99 times in comparison with the standard CGP.

Anglický abstrakt

The aim of this paper is to accelerate the task of evolutionary image filter design using coevolution of candidate filters and training vectors subsets. Two coevolutionary methods are implemented and compared for this task in the framework of Cartesian Genetic Programming (CGP). Experimental results show that only 15-20 % of original test vectors are needed to find an image filter which provides the same quality of filtering as the best filter evolved using the standard CGP which utilizes the whole training set. Moreover, the median time of evolution was reduced 2.99 times in comparison with the standard CGP.

BibTex


@article{BUT96949,
  author="Michaela {Drahošová} and Lukáš {Sekanina}",
  title="Acceleration of Evolutionary Image Filter Design Using Coevolution in Cartesian GP",
  annote="The aim of this paper is to accelerate the task of evolutionary image filter
design using coevolution of candidate filters and training vectors subsets. Two
coevolutionary methods are implemented and compared for this task in the
framework of Cartesian Genetic Programming (CGP). Experimental results show that
only 15-20 % of original test vectors are needed to find an image filter which
provides the same quality of filtering as the best filter evolved using the
standard CGP which utilizes the whole training set. Moreover, the median time of
evolution was reduced 2.99 times in comparison with the standard CGP.",
  address="Springer Verlag",
  booktitle="The 12th International Conference on Parallel Problem Solving from Nature",
  chapter="96949",
  doi="10.1007/978-3-642-32937-1_17",
  edition="NEUVEDEN",
  howpublished="print",
  institution="Springer Verlag",
  number="7491",
  volume="2012",
  year="2012",
  month="july",
  pages="163--172",
  publisher="Springer Verlag",
  type="journal article - other"
}