Graph Cut Based Multiple Interactive Image Segmentation for Medical Applications
Mamatha S K,  Krishnappa H K
Segmentation of medical images could be a basic and difficult task in the field of medical applications. There are several existing medical image segmentation approaches, among these existing graph cut methods are comparatively new and has good features in medical applications. Segmenting clinical images could be effortful, long method, slow process and automatic image segmentation method typically contain errors and imperfections. Interactive image segmentation is associate rising technology within the areas of image processing, computer vision and medical field. Graph cut based Multiple interactive segmentation is presented is in three steps. Initially, nodes representing pixels of image area connected to their k-nearest neighbors to cover foreground image. Within the second step, energy function of graph is employed to improve the segmentation on the object borders to hide background information set. Third step extracting user interacted object from image set. This paper also investigated issues of previous methods like mean shift segmentation, watershed technique and automatic graph cut based image segmentation. As a result, the graph cut method cut method shows higher performance than previous methods.
Keywords- Interactive segmentation, Automatic graph cut, Mean shift segmentation
Cite this Article
Mamatha S K,  Krishnappa H K,   "Graph Cut Based Multiple Interactive Image Segmentation for Medical Applications"
, International Journal of Engineering Development and Research (IJEDR), ISSN:2321-9939, Volume.7, Issue 3, pp.567-571, September 2019, Available at :http://www.ijedr.org/papers/IJEDR1903098.pdf