A Reformative Gradient Vector Flow Based on Beltrami Flow

Abstract

The gradient vector flow (GVF) snake shows high performance at concavity convergence and initialization insensitivity, but the two components of GVF field are treated isolatedly during diffusion, this leads to the failure of GVF snake at deep and narrow concavity convergence and weak edge preserving. In this study, a novel external force for active contours is proposed that couples the two components during diffusion by generalizing the Laplacian operator from flat space to manifold. The specific operator is Beltrami operator. This proposed method has been assessed on synthetic and real images; experimental results show that the new method behaves similarly to the GVF snake in terms of capture range enlarging, initialization insensitivity, U-shape concavity convergence, while provides much better results than GVF snake for noise robustness, narrow and deep concavity convergence and weak edge preserving.