Abstract: Aiming at the problem of multi-object image segmentation, a novel image segmentation method is proposed based on mutual information and Chan-Vese model, which integrates the idea of multilevel segmentation and introduces the concept of mutual information in information theory instead of gray average deviation to decide whether the segmentation has been wholly completed. The results of experiments indicate its validity and effectiveness in segmentation of multi-object image and the objects with weak boundaries.

Key words: image segmentation, level set, Chan-Vese(C-V)model, multiphase segmentation, mutual information

摘要： 针对多目标物体图像的分割问题，该文在Chan-Vese模型(C-V模型)的基础上，提出了基于互信息和Chan-Vese模型的图像分割方法。该方法结合多级分割的思想，引入了信息论中互信息的概念，替代多级分割中的灰度平均方差，将互信息量作为判断分割是否完成的标准。实验结果表明，该方法能够有效地解决多目标物体图像以及弱边界物体的分割问题。

关键词: 图像分割, 水平集, Chan-Vese(C-V)模型, 多相分割, 互信息

CLC Number: 

• TP391.41