Active Contour Model Coupling with Backward Diffusion for Medical Image Segmentation

Active contour models are very useful for image segmentation, but it is not true for images with intensity inhomogeneities which often occur in medical images. The reason is that the weak edge informations are disturbed by the intensity inhomogeneities, and the segmentation will be success if we enhance the edges. In order to overcome the difficulties caused by intensity inhomogeneities, we propose a region-based active contour model that coupling with backward diffusion which has the ability of edge enhancement for segmentation. In our model we replace the data term of piecewise constant approximation in CCV (Convex Chan-Vese) model with backward diffusion model to realize the alternating minimization of parameters of active contour evolution. Finally, the fast Split Bregman algorithm of the proposed coupling model is designed for the segmentation implementation. The performance of our method is demonstrated through numerical experiments of some medical image segmentations.