Using the Split Bregman Algorithm to Solve the Self-repelling Snakes Model

Abstract Preserving contour topology during image segmentation is useful in many practical scenarios. By keeping the contours isomorphic, it possible to prevent over-segmentation and under-segmentation, as well adhere given topologies. The Self-repelling Snakes model (SR) a variational that preserves by combining non-local repulsion term with geodesic active model. SR traditionally solved using additive operator splitting (AOS) scheme. In our paper, we propose an alternative solution Split Bregman method. Our algorithm breaks problem down into simpler sub-problems use lower-order evolution equations simple projection scheme rather than re-initialization. can be via fast Fourier transform or approximate soft thresholding formula which maintains stability, shortening convergence time, reduces memory requirement. AOS algorithms are compared theoretically experimentally.