一种基于图割优化的局部水平集图像分割方法

Local Level Set Image Segmentation Based on Graph Cut Optimization Method

针对传统水平集方法中运算效率较低的问题,提出一种基于图割优化的融合局部统计信息的水平集图像分割方法。研究表明,图割优化算法对于改善传统水平集能量函数最小化求解效率具有一定的优势,将图割优化算法引入局部统计分析模型能量函数的数值化,根据该问题与局部统计分析模型能量函数的等价性,提出一种基于局部统计分析模型映射到图割模型中各边权重的构造方法,利用图割优化中的连续最大流最小割模型,实现了局部统计模型能量函数的最小化求解。实验结果证明,该方法能够保持局部统计模型对灰度不均匀图像的敏感性,明显提高了计算效率。

To improve the low efficiency in the traditional level set method for image segmentation,we propose to use graph-cut optimization algorithm to optimize local statistical analysis model of level set method. Researches show that,graph-cut has an advantage in improving the minimized energy function efficiency of traditional level set. Meanwhile,this paper introduces the numerical method of local statistical analysis measurement with graph-cut optimization and discusses the similarity between this problem and the energy function of local statistical analysis model. We convert the level set energy function to the frame of graph-cut whose energy function can be efficiently minimized by max-flow min-cut algorithm. The experiments demonstrate that the proposed method can achieve satisfactory segmentation for images with intensity inhomogeneity as well as very high efficiency.