2D Image Segmentation Through the One Dimensional Mumford-Shah Functional

Image segmentation is a complex task, since it involves the manipulation of different kind of objects such as functions, domains in R and curves. In this work we propose a two-dimensional image segmentation method, using the Mumford and Shah’s one dimensional technique, performing vertical and horizontal sweeps. This is in contrast to the technique that involves the generalization of the Mumford and Shah functional theory for 2D images.