Image segmentation using genetic algorithm for four gray classes

Image segmentation is a technique of image analysis which gives information about the different homogeneous regions in given image. The segmented region may be a complete object or part of it. M. Yu. et al. [5] have developed method for segmentation of images containing two gray classes. In the proposed method, we have extended that method for four gray classes. Generally, in GA initial populations are generated randomly. The fitness function is used to evaluate the solutions and the fittest solutions are selected as parents for producing offspring´s that form the next generations. Morphological operations are used in reproduction step of GA. After several generations, populations are evolved to get the near optimal results. We present the experimental result, which demonstrates the segmentation of the image into four classes using GA.