Prof. Ing. Lukáš Sekanina, Ph.D.
DRAHOŠOVÁ Michaela and SEKANINA Lukáš. Acceleration of Evolutionary Image Filter Design Using Coevolution in Cartesian GP. 8th Doctoral Workshop on Mathematical and Engineering Methods in Computer Science. Brno: Masaryk University, 2012. ISBN 9788087342152.  Publication language:  english 

Original title:  Acceleration of Evolutionary Image Filter Design Using Coevolution in Cartesian GP 

Title (cs):  Akcelerace evolučního návrhu obrazových filtrů s použitím koevoluce v kartézském GP 

Pages:  1 

Book:  8th Doctoral Workshop on Mathematical and Engineering Methods in Computer Science 

Conference:  MEMICS'12  8th Doctoral Workshop on Mathematical and Engineering Methods in Computer Science 

Place:  Brno, CZ 

Year:  2012 

ISBN:  9788087342152 

Publisher:  Masaryk University 

Keywords 

Cartesian genetic programming, coevolution, fitness modeling, image filter design. 
Annotation 

This contribution is based on the paper Acceleration of Evolutionary Image Filter Design Using Coevolution in Cartesian GP, that has been pubished in The 12th International Conference on Parallel Problem Solving from Nature, LNCS 7491, Berlin, DE, Springer, 2012, p. 163172, ISBN 9783642329364. 
Abstract 

Evolutionary design based on genetic programming is a very computationally intensive design method. It also holds for the evolutionary design of image filters which has been performed by Cartesian Genetic Programming (CGP). The most time consuming procedure is the fitness calculation where tens of thousands of pixels in training set (the socalled set of target objective vectors, TOVs) have to be evaluated in order to obtain a single fitness value. A single run is typically finished after 200 thousands candidate filter evaluations.
In this work, we propose to employ a coevolutionary algorithm based on coevolution of candidate filters and TOVs subsets running on an ordinary processor to accelerate the image filter evolution. We use two populations: (1) population of candidate filters evolved on principles of CGP and (2) population of TOVs subsets evolved using a simple genetic algorithm (GA); each population is running in one thread (i.e. we use a twothread model). The aim of this type of coevolution is to allow both candidate programs and TOVs subsets to improve each other automatically until a satisfactory problem solution is found. We compared two approaches to fitness calculation of TOVs subset  the first one is based on the fitness predictor concept (CFP) and the second approach exploits the competitive coevolution scheme (CC). The saltandpepper noise filters are designed using standard CGP and coevoltuionary CGP.
Experimental results show that only 1520% of original training 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 (the execution time is the sum for both threads) in comparison to the standard CGP running in one thread. 
