An improved C-V level set model

摘要

The traditional geodesic active model with edge(C-V Model) can deal with topological change automatically. But C-V Model is sensitive to noise and inaccurate positioning of the edges in an image. This paper proposes a C-V model based on an improved Gaussian Laplace operator (LOG). The improved Gaussian Laplace operator can enhance the accuracy of the edge detection and the noise immunity. Experiments are showed that the proposed algorithm can achieve good segmentation effect.