A Comparative Review of Image Segmentation Techniques

In today’s computer science image segmentation has become the main focus of research as a universal algorithm for segmentation of images. There is no one method in the world for segmentation which is good for all type of images and produces the same quality of results always for all types of images. Due to these types of challenges the sector image segmentation always remains a focus of main research under the computer science e and image processing algorithm’s it is considered as still a pending problem for the world of computers. Despite of these facts the image segmentation can still has given more methods for different kinds of problems the any other techniques under the image segmentation. This methodology is used to extract particular part of images as a separate identity itself. Unsupervised image segmentation is also a incomplete data problem because of the fact that model parameters and class labels are still unknown. There are many of the algorithms and basic methods developed during the time frame under image segmentation but need for new methods are still in need according to changing needs and advancement of technology. In this paper, we have analyzed and reviewed EMHMRF, Gaussian mixture model GMM-HMRF and K-GMM-EMM method and compare them on the basis of Time and energy activation graph.