A variable-domain approach for image segmentation based on statistical models

In this paper, a novel variable-domain approach to curve evolution for image segmentation was proposed, being based on a statistical active contour model using level sets. The essential idea is to re-define the computing domain in image repeatedly, by separating the segmentation procedure into several individual phases, for images composed of an infinite number of regions. By our algorithm, the work can be done automatically without manual intervention. Moreover, the accuracy and rapidity can be enhanced effectively for the objects with complicated topology.