Active Contours and Mumford—Shah Segmentation Based on Level Sets
Nassir, Hanaa Mohsin Ahmed, 刘重庆 · 上海交通大学学报：英文版 · 2003
This paper is to detect regions (objects) boundaries, also to isolate and extract individual componentsfrom a medical image. This can be done using an active contours to detect regions in a given image, based on tech-niques of curve evolution, Mumford-Shah functional for segmentation and level sets. The paper classified the im-ages into different intensity regions based on Markov random field, then detected regions whose boundaries are notnecessarily defined by gradient by minimizing an energy of Mumford-Shah functional for segmentation which can beseen as a particular case of the minimal partition problem. In the level set formulation, the problem becomes amean-curvature flow like evolving the active contour, which will stop on the desired boundary. The stopping termdoes not depend on the gradient of the image, as in the classical active contour and the initial curve of level set canbe anywhere in the image, and interior contours are automatically detected. The final image segmentation is oneclosed boundary per actual region in the image.