Automatic Image Segmentation using Threshold Based Parameter and Genetic Algorithm

Image segmentation is a significant step in any computer vision system. It can be explore as a process of isolating objects of interest from the rest of the scene. In this paper, we propose a new threshold based information parameter for image segmentation through Genetic Algorithm. We used Genetic Algorithm because it is able to determine the optimal number of regions of segmentation. The developed method uses an evolutionary criteria and produce best segment selection. In proposed method, Genetic Algorithm is used in order to determine evolutionary best segmented image based on a new information based parameter via threshold. This method is implemented only on gray level images and in last section we have shown our experimental results by applying the proposed method on different gray images. We also review some of the existing Image Segmentation techniques.