Improving search performance of linear genetic programming based image recognition program synthesis by redundancy-removed recombination

This paper propose a new recombination method, named redundancy-removed recombination, for linear genetic programming based image recognition program synthesis. The redundancy-removed recombination produces an offspring (by conventional crossover or mutation), and then adopts a canonical transformation to convert the offspring into its canonical form, in which it can be verified whether it has been evolved before (redundant). If the offspring is redundant, it is prohibited and recombination is repeated until non-redundant offspring, which has never be born in the evolutionary search, is produced. Experimental results show that the use of the redundancy-removed recombination improved the performance of evolutionary search; it converged to the global optimum faster than the use of conventional recombinations. Also we found that the redundancy-removed recombination can construct longer programs and concentrate on those areas, whereas the conventional ones cannot.

[1]  Kenneth de Jong,et al.  Evolutionary computation: a unified approach , 2007, GECCO.

[2]  Colin R. Reeves,et al.  Evolutionary computation: a unified approach , 2007, Genetic Programming and Evolvable Machines.

[3]  Tomoharu Nagao,et al.  Automatic construction of tree-structural image transformations using genetic programming , 1999, Proceedings 10th International Conference on Image Analysis and Processing.

[4]  Mark E. Roberts,et al.  A Multistage Approach to Cooperatively Coevolving Feature Construction and Object Detection , 2005, EvoWorkshops.

[5]  Bir Bhanu,et al.  Multiclass Object Recognition Based on Texture Linear Genetic Programming , 2007, EvoWorkshops.

[6]  Wolfgang Banzhaf,et al.  A comparison of linear genetic programming and neural networks in medical data mining , 2001, IEEE Trans. Evol. Comput..

[7]  Tomoharu Nagao,et al.  GENETIC IMAGE NETWORK ( GIN ) : AUTOMATICALLY CONSTRUCTION OF IMAGE PROCESSING ALGORITHM , 2006 .

[8]  Mark E. Roberts,et al.  Cooperative Coevolution of Image Feature Construction and Object Detection , 2004, PPSN.

[9]  W. Spears,et al.  On the Virtues of Parameterized Uniform Crossover , 1995 .

[10]  Zbigniew Michalewicz,et al.  Genetic Algorithms + Data Structures = Evolution Programs , 1996, Springer Berlin Heidelberg.

[11]  Victor Ciesielski,et al.  Linear genetic programming , 2008, Genetic Programming and Evolvable Machines.

[12]  Astro Teller,et al.  PADO: a new learning architecture for object recognition , 1997 .

[13]  Manuela M. Veloso,et al.  A Contolled Experiment: Evolution for Learning Difficult Image Classification , 1995, EPIA.

[14]  James R. Levenick.  Inserting Introns Improves Genetic Algorithm Success Rate: Taking a Cue from Biology , 1991, ICGA.

[15]  Krzysztof Krawiec,et al.  Visual learning by coevolutionary feature synthesis , 2005, IEEE Transactions on Systems, Man, and Cybernetics, Part B (Cybernetics).

[16]  Krzysztof Krawiec,et al.  Visual Learning by Evolutionary and Coevolutionary Feature Synthesis , 2007, IEEE Transactions on Evolutionary Computation.

[17]  Krzysztof Krawiec,et al.  Generative learning of visual concepts using multiobjective genetic programming , 2007, Pattern Recognit. Lett..

[18]  Tomoharu Nagao,et al.  Automatic construction of image transformation processes using genetic algorithm , 1996, Proceedings of 3rd IEEE International Conference on Image Processing.