Comparison Between New Operators Of Genetic Algorithm To Implementation Face Image

H. Karruffa, Shahla

	Comparison Between New Operators Of Genetic Algorithm To Implementation Face Image
Tikrit Journal of Pure Science
Article 1, Volume 16, Issue 2, 2011, Pages 87-99
Author
Shahla H. Karruffa
Abstract
Genetic algorithm (GA) are adaptive, robust, efficient, and global search methods, suitable in situations where the search space is large. They optimize a fitness function, corresponding to the preference criterion to arrive at an optimal solution using certain genetic operators. GA are able to produce valuable information for an enterprise (Like Face images), it is only when the GA are implemented thoroughly and by a resource qualified to utilize the data that an enterprise can fully capitalize on the benefits. I have dealt with in this research with basic processes in the GA like selection, crossover and mutation. The concentration was on mutation being has an important on the results. New five methods were suggested for mutation process for implementing face image, and these methods are: (1- Separation genetic operator, 2- Annealing genetic operator, 3- Swap genetic operator, 4- Block-swap genetic operator, 5- Suffle genetic operator). Later, the comparing and finding the best method among them, where the research includes dividing the work into three stages. In the first stage when implementation number is low, while the second one when implementation number is intermediate ,but the third stage when implementation number is high. The program was done in Matlab language (Version 6.5) for performing the work and dealing with each image alone and the images were displayed and drawing the diagram sketches for the resulted image.

Statistics Article View: 61 PDF Download: 29