A Fast Convergent Snake Model Based on Distance Transform

We present a novel and fast convergent Snake model based on the distance transform. By the method, we find the points in the disconvergent area based on the property of external forces in the concavity; Secondly, we modify the direction and magnitude of the external forces at these points to make the convergence fast. Our algorithm is better than the GVF algorithm in terms of convergence speed and the result for deep boundary concavities. Besides, our model can be also extended to Snake model based on GVF.