Omaima N. Ahmad AL-Allaf
© Springer International Publishing Switzerland 2015
K. Arai et al. (eds.), Intelligent Systems in Science and Information 2014,
Studies in Computational Intelligence 591, DOI 10.1007/978-3-319-14654-6_25
Abstract
This chapter presents how to use parallel computing approaches from
MATLAB Parallel Computing Toolbox to implement genetic algorithm for fractal
image compression. These approaches are: ParFor, CoDistributor and Parallel
Cluster. This is done to decrease processing time as possible as and maintaining
reconstructed image quality. Many experiments were executed with comparisons
between the three approaches. The experimental results showed that decreasing
the GA population size and increasing number of workers used for the three parallel
computing approaches can reduce the compression time. Best results obtained
from implementing parallel approaches with 6 workers and 150 population size.
The execution speed reached 4, CR reached 90.97 % and PSNR reached 34.98 db.
At the same time, best results obtained from Parallel Cluster approach and then
from CoDistributor approach.