Active contours driven by regularised gradient flux flows for image segmentation

A novel active contour model driven by regularised gradient flux flows is presented for image segmentation, which achieves an accurate result because the zero crossings of the image Laplacian are reached at the object boundary when minimising the gradient flux flows. Furthermore, the Laplacian of the image is regularised with an anisotropic diffusion term, which can not only reduce noise but also preserve edge information. Experiments on several synthetic and real images validate the proposed method with promising results.