期刊名称：International Journal of Advanced Research in Computer Engineering & Technology (IJARCET)
印刷版ISSN：2278-1323
出版年度：2013
卷号：2
期号：4
页码：1483-1487
出版社：Shri Pannalal Research Institute of Technolgy
摘要：Image Segmentation is a decomposition of scene into its components. It is a key step in analysis. Edge, point, line, boundary texture and region detection are the various forms of image segmentation. Various technologies for image segmentation are there like thresholding, cluster based, edge based, region based and watershed segmentation. Two of the main image segmentation techniques thresholding and region growing are highly in use for image segmentation. Image segmentation by region growing method is robust fast and very easy to implement, but it suffers from: the thresholding problem, initialization, and sensitivity to noise. OTSU method of thresholding is also used for image segmentation but it also suffers from thresholding problems. Genetic algorithms are particular methods for optimizing functions; they have a great ability to find the global optimum of a problem. Here I proposed a genetic algorithm which provides the better solution than region growing and OTSU methods for the image segmentation. In proposed algorithm we will see that we get better peak signal to noise ratio and maximum absolute error comparatively than region growing and OTSU.
关键词：genetic algorithm; Image Segmentation; ; OTSU; region growing
