Improvement of image segmentation based on GVF model
基于GVF模型的图像分割方法的改进

Based on the analysis of the force field of gradient vector flow (GVF) model, the force vectors of GVF in the diffusion equations were normalized, which made the external forces irrelevant of the distance between the points of the contour and the boundary of object. Thus this method can solve deep boundary concavities. At the same time, by introducing direction vector into the GVF model, in the region specified by this vector, the force fields of the objects not wanted can be shielded, and the interested objects can be segmented more precisely.