Snake的初衷是为了进行图像分割，但它对初始位置过于敏感，且不能处理拓扑结构改变的问题。初始位

置的敏感性可以用遗传算法来克服，因为它是一种全局优化算法，且有良好的数值稳定性。为了更精确地进行图

像分割，本文提出了一种基于遗传算法的双T—Snake模型图像分割方法，它将双T—Snake模型解作为遗传算法的搜

索空间，这既继承了T—Snake模型的拓扑改变能力，又加快了遗传算法的收敛速度。由于它利用遗传算法的全局优

化性能，克服了Snake轮廓局部极小化的缺陷，从而可得到对目标的更精确的分割。将其应用于左心室MRI图像的分割，取得了较好的效果。-Snake s original intention was to carry out image segmentation, but it was too sensitive to initial position and can not deal with the issue of topology change. Initial position

The sensitivity of home can be used to overcome the genetic algorithm because it is a global optimization algorithm, and have good numerical stability. In order to more accurately map

Like segmentation, this paper presents a genetic algorithm based on dual-T-Snake model for image segmentation method, it will double-T-Snake model solution found as a genetic algorithm

Cable space, which not only inherited the T-Snake model the ability to change the topology, but also speed up the convergence rate of genetic algorithm. It uses genetic algorithms as a result of the overall excellent

Of performance, to overcome the local minimum of Snake contour deficiencies, which can be more precise on the target partition. Will be applied to MRI images of left ventricle

Segmentation, and achieved good results.