Image Segmentation with GVF Snake and Corner Detection

Abstract

Gradient vector flow (GVF) snake model is used widely in image segmentation and computer vision. GVF snake has larger capture range and stronger convergence ability to boundary concavities than traditional snake. However in the energy minimization process some corner points canpsilat be found. Because of this the object boundary is not accurate. In this paper, a new image segmentation algorithm based on Susan approach and GVF Snake model is proposed. First we check corner points at the edge using Susan approach and mark them as energy minimization points, then use GVF Snake model to capture object boundary after set initial snake curve. Experiments indicate that the new algorithm can improve GVF snake modelpsilas precision to capture the boundary with sharp-angled corner.