National Repository of Grey Literature 4 records found  Search took 0.00 seconds. 
Evolutionary Design Using Rewriting Systems
Hýsek, Jiří ; Jaroš, Jiří (referee) ; Bidlo, Michal (advisor)
This work provides an introduction to the evolutionary algorithms and evolutionary design. It also describes disadvantages of direct encoding of a genotype to a phenotype and a method of nontrivial encoding which can solve these problems. We are particularly talking about the problems of the scalability of evolved solutions. We discuss a possible solution of described problem - a nontrivial genotype-phenotype mapping called development. This technique is demonstrated on an evolutionary design of a sequence of rewriting rules which is able to construct arbitrarily large sorting networks.
Coupling of images
Nováček, Petr ; Petyovský, Petr (referee) ; Richter, Miloslav (advisor)
The thesis deals with coupling of non-overlapping images with square parts. The target of this thesis is a design of an algorithm for coupling of images based on color similarity of individual parts edges. A genetic algorithm has been chosen for coupling of images. The first part of this thesis lists the theory of image processing and genethic algorithms. The second part deals with applied procedures for image processing and with design of genetic algorithm. The last part is devoted to an evaluation of results of the applied algorithms.
Coupling of images
Nováček, Petr ; Petyovský, Petr (referee) ; Richter, Miloslav (advisor)
The thesis deals with coupling of non-overlapping images with square parts. The target of this thesis is a design of an algorithm for coupling of images based on color similarity of individual parts edges. A genetic algorithm has been chosen for coupling of images. The first part of this thesis lists the theory of image processing and genethic algorithms. The second part deals with applied procedures for image processing and with design of genetic algorithm. The last part is devoted to an evaluation of results of the applied algorithms.
Evolutionary Design Using Rewriting Systems
Hýsek, Jiří ; Jaroš, Jiří (referee) ; Bidlo, Michal (advisor)
This work provides an introduction to the evolutionary algorithms and evolutionary design. It also describes disadvantages of direct encoding of a genotype to a phenotype and a method of nontrivial encoding which can solve these problems. We are particularly talking about the problems of the scalability of evolved solutions. We discuss a possible solution of described problem - a nontrivial genotype-phenotype mapping called development. This technique is demonstrated on an evolutionary design of a sequence of rewriting rules which is able to construct arbitrarily large sorting networks.

Interested in being notified about new results for this query?
Subscribe to the RSS feed.