摘要
水平集方法已广泛应用于医学图像分割中,该方法将界面看成高一维空间中的某一函数覬(称为水平集函数)的零水平集,同时界面的演化也扩充到高一维的空间中。其核心思想是利用水平集理论求解能量泛函的最小值,即当能量达到最小值时的曲线位置就是目标轮廓所在;有效解决曲线演化过程中的拓扑变化问题。介绍水平集发展过程中几个经典模型的基本思想,并通过大量实验证明该方法在医学图像分割中的适用性及有效性。
Evolution of the level set method has been widely applied in medical image segmentation, which will screen as a higher-dimensional space in a certain function （called the level set function） of the zero level set, while also expanding the interface to a higher-dimen- sional space. The core idea is to put the level set on the mathematical theory minimum energy functional solution process, when the curve reaches a minimum energy position is where the target contour lies, effectively solving the problem of topology change in the evolu- tion of the curve which has no proper algorithm to solve previously. Describes several classical models＂ basic idea of the level set devel- opment and by a large number of experiments proves the applicabilitv and effectiveness of this method in medical image segmentation.
关键词
水平集
医学图像分割
能量泛函
曲线演化
Level Set
Medical Image Segmentation
Minimum Energy Functional
Curve Evolution
