A fractional-order regulatory CV model for brain MR image segmentation

In this paper, we introduce fractional derivative into CV level set model for image segmentation. Specifically, the first-order gradient operator in the CV level set model is generalized to fractional-order gradient by energy formulation regulation, which considers the nonlinear protecting capability of fractional-order derivative for texture and lower frequency features of images. The corresponding fractional Euler-Lagrange equation is given for level set evolution, and then the numerical algorithm is analyzed. The novel model has been validated on real and simulated brain MR images, with desirable performance in the presence of intensity inhomogeneity, compared with the traditional CV level set model.