State-of-the-Art Level Set Models and Their Performances in Image Segmentation: A Decade Review

In modern days, image segmentation is one of the most important processing step in the field of computer vision and image processing. It helps to identify object, reconstruct shape, classify and estimate volume of an object. In the last few decades, many algorithms have been developed to eradicate the various segmentation problems such as weak edge detection, inhomogeneous image segmentation, accurate object shape identification and classification. Among them, one of the popular active contour models namely level set model is extensively used to eliminate the problem of topological changes during curve evolution. Earlier, the active contour models were unable to deal with sudden topological changes which led to poor segmentation results. Thus, the paper investigates several level set models in various applications of modern imaging. Therefore, it is necessary to understand the formulation of various level set models with their characteristics before applying them to solve the segmentation problem. In this paper, authors have extensively studied the formulation of various level set models and their application in different types of images. Further, the authors have discussed their contributions to level set framework and open research challenges for researchers.