Genetic Algorithm (GAs) was introduced by Holland in early 1970s.
In contrastwith evolution policies and programming, Holland’s unique goal was not to designalgorithms to solve specific problems, but rather to formally study the phenomenon of adaptation as it occurs in nature. He was looking to develop methods in which thenatural adaptation can be introduced into computer systems. Holland’s 1975 bookAdaptation in Natural and Artificial Systems 5 presented the Genetic Algorithm asa generalization of biological evolution and gave a theoretical outline for adaptationunder the GA. Holland’s GA is a method for moving from one population of “chromosomes” (e.
g., group of ones and zeros called bits) to a new population by usingwhat is called natural selection together with the genetics operators of crossover ,and mutation . The chromosome consists of genes such that each gene is an instanceof a particular allele with value of 0 or 1. The selection operator picks the chromosomesin the population in order to reproduce, and on in general the best chromosomesreproduce more than the less fit ones.
Mutation randomly changes the allelevalues of some random selected locations in the chromosome. The chromosomes ina GA population typically take the form of bit strings or integer number. Each genein the chromosome has two possible alleles: 0 and 1 for bit strings or any value Z. Incase of images it is an integer value that ranges between 0 and 2Rbits where Rbits isthe radiometric resolution. Each chromosome can be assumed of as a position in thesearch space for a candidate solutions. The GA processes population s of chromosomes, successively replacing one such population with another. The GA most oftenrequires a fitness function that assigns a score (fitness ) to each chromosome in thecurrent population . The fitness of a chromosome depends on how well that chromosomesolves the problem at hand.
The following pseudo code shows how GA works: