Image Segmentation on Colonies Images by A Combined Algorithm of Simulated Annealing and Genetic Algorithm

For the segmentation of several typical classes of colony images, there are large numbers of applications. This paper describes a combined algorithm for colony image segmentation. The problem of image segmentation is treated as one of combinatorial optimization. The simulated annealing (SA)-based image segmentation technique is seeing to suffer from several limitations. The search procedure of SA is fairly localized, preventing them from exploring the same diversity of solutions. Although genetic algorithm (GA) has an excellent capability of global researching, its capability of hill-climbing is weak. This combined algorithm may be advantageous in combining the advantages of both GA and SA procedures while alleviating their individual shortcomings. Experiments show that the combined algorithm provides a useful method for colony image segmentation, and the whole image segmentation process time is several time short more than traditional approaches.