A New Segmentation Model--Combined Mumford-Shah Model with Narrow Band

A segmentation model that combines the Mumford-Shah (M-S) model and narrow band scheme of level set is presented. The M-S model is a desirable model for image segmentation, but computationally time-consuming. This paper introduces a fast segmentation model, which combines the M-S model and narrow band scheme using new initialization method. The new initialization method is based on fast marching method, and the computing time is O(n). In each iteration step, the new segmentation model only deals with the data in narrow band instead of the whole image. Comparing M-S model and new narrow band M-S, experiments show that the two models can obtain almost the same segmentation result, but the computing time of new narrow band M-S model is much less than M-S model.