一种基于区域填充的C-V模型图像分割方法

传统的C-V模型对于包含有多灰度级对象或子对象的图像难以实现准确分割.为解决这一问题,提出了一种基于区域填充的C-V模型图像分割方法,通过对象域或背景域的填充,将主对象转变为背景,从而弱对象或子对象成为主对象,再通过C-V模型实现分割.数值实验表明,该方法能够将弱对象有效分割,而且对于复杂对象的分割较C-V模型更为准确.

Image Segmentation Method of C-V Model Based on Region Painting

The traditional C-V model could not accurately segment an image including multi-gray level objects or sub-objects. In order to solve these problems, an image segmentation method based on C-V model and region painting was proposed. By object or background region painting, the main objects were changed to be part of background, so the weak object or sub-object became main object and could be detected by C-V model. Experiments show the proposed method can effectively segment the weak object that can not be detected originally, and is more accurate in segmentation of complicated object than C-V model.