Publication detail
Back-Propagation and Genetic Algorithms for Image Processing
ŠŤASTNÝ, J. ŠKORPIL, V.
Czech title
Použití algoritmu Back-propagation a genetického algoritmu pro obrazovou analýzu
English title
Back-Propagation and Genetic Algorithms for Image Processing
Type
conference paper
Language
en
Original abstract
The contibution describes the problem of the image processing by using artificial neural networks. The Multi Layer Perceptron and Radial Basis neural networks were used. We compared results obtained by a using of different learning algorithms, the classical back-propagation algorithm and the genetic algorithm. The real technological scene for image processing was simulated with digitization of two-dimensional pictures.
Czech abstract
Příspěvek popisuje problematiku obrazové ananlýzy při aplikaci umělých neuronových sítí. Byly použity neuronové sítě MLP a RBFN. Byly porovnány výsledky dosažené aplikací algiritmu Back -propagation a genetického algoritmu. Technologická scéna pro obrazovou analýzu byla simulována digitalizací dvojrozměrných obrazů.
English abstract
The contibution describes the problem of the image processing by using artificial neural networks. The Multi Layer Perceptron and Radial Basis neural networks were used. We compared results obtained by a using of different learning algorithms, the classical back-propagation algorithm and the genetic algorithm. The real technological scene for image processing was simulated with digitization of two-dimensional pictures.
Keywords in English
Image Processing, Genetic Algorithm, Back-Propagation Algorithm
RIV year
2005
Released
03.11.2005
Publisher
University Politehnica of Bucharest
Location
Bukurešť, Rumunsko
ISBN
973-27-1254-6
Book
The 4th International conference on Advanced Manufacturing Technologies ICAMaT 2005
Pages count
4
BIBTEX
@inproceedings{BUT18014,
author="Jiří {Šťastný} and Vladislav {Škorpil},
title="Back-Propagation and Genetic Algorithms for Image Processing",
booktitle="The 4th International conference on Advanced Manufacturing Technologies ICAMaT 2005",
year="2005",
month="November",
publisher="University Politehnica of Bucharest",
address="Bukurešť, Rumunsko",
isbn="973-27-1254-6"
}