Genetic Algorithms: Colour Image Segmentation Literature Review

Image segmentation is a crucial problem in image processing and can determine the final outcome of many image processing tasks. Lots of research has gone into segmentation of monotone images, however, recently more research has gone into segmenting colour images. Genetic algorithms have been shown to be a viable method to segment an image. The island model has been shown to be the most effective model for parallel genetic algorithms. However, little research has gone into developing a PGA for a Grid computing environment, which consists of heterogeneous, nondedicated resources. This paper presents some of the key segmentation techniques that have been developed. It then goes into genetic algorithms and how they have been parallelised. It finally looks at the little research that has been done on Grid-based GAs.