In classic clone selection algorithm, all solutions (individuals) are cloned and mutated in the same way, and most individuals to participate in the evolution are those who have a higher degree of affinity, the lower ones have less opportunity to participate in the evolution, it can lead to the diversities disorder easily, and the premature convergence can also happen.
The proposed Distributed Clone Selection Algorithm combines different mutation strategies in the uniform evolutionary framework to utilize the strength of
local and global search method as well as to avoid the premature convergence.
The proposed algorithm is evaluated and validated by its application in image enhancement experiments.