改进的分级多相图像分割模型及其快速实现

Improved Hierarchical Multiphase Image Segmentation Model and Its Fast Implementation

提出一种基于分级C-V模型的改进的快速图像分割算法。针对现有的多相水平集图像分割算法存在的问题,本文从曲线演化方程的平均曲率项、水平集函数Φ的狄拉克(Dirac)函数δ(Φ)等方面进行改进,并引入了一个非线性扩散方程对图像进行预处理,从而优化组合了分级C-V模型的全局特性。实验结果表明,改进的图像分割模型不仅保留了原有方法的优势,而且提高了对多目标图像分割算法的速度与精度,同时也可以有效解决具有弱边界物体的分割问题。

An improved fast algorithm is presented for the image segmentation based on the hierarchical C-V model. To overcome limitations of these multiphase level-set methods, this pa- per modifies the mean curvature and Dirac function in the curve evolution equations, and intro- duces a nonlinear diffusion equation as a preprocessing step on the image. Thus, the global characteristics of the hierarchical C-V model are optimized. Experimental results show that the method maintains the advantages of the hierarchical C-V model and improves the speeds and accuracies of the segmentation of multi-object image and objects with weak boundaries.