中文摘要：

水平集方法（LSM）图像分割的本质是求解一个随时间变化的偏微分方程,而使用变分法求解此水平集方程（LSE）往往要耗费过多的计算时间。为了减少算法的运行时间,提出了一种快速水平集图像分割算法。该算法在模糊聚类水平集方法（FCM-LSM）的基础上使用高斯混合模型（GMM）改造其隶属度损失函数,并利用离散网格Boltzmann方法（LBM）求解水平集方程。实验结果表明：本文提出的算法无论是在执行效率上还是在分割效果上都优于传统方法,证明了算法的可行性。

英文摘要：

The level set method（LSM）for image segmentation is used to solve a time-varying partial differential equation,which may require too much calculation time by using the calculus of variation method.Aiming at the problem,this paper proposes a fast level set method for image segmentation.Basing on fuzzy clustering level set method（FCM-LSM）,the proposed algorithm utilizes Gaussian mixture models（GMM）to modify the membership loss function.And then,the lattice Boltamann method（LBM）is used to solve the level set equation.Experimental results show that the proposed algorithm is effective in efficiency and segmentation results.