MRI分割的几何活动轮廓模型的半隐式AOS迭代方法

Semi-Implicit AOS Schemes for MRI Segmentation Geodesic Active Contour Model

几何活动轮廓模型以其计算稳定等优点,被广泛用于图像分割。给出一种基于AOS(additiveoperatorsplitting)算法的快速活动轮廓模型,将AOS方法与几何活动轮廓模型相结合构造一种半隐式迭代模型,它对迭代步长没有限制,从而可以选择较大的迭代步长,提高了模型的时间效率,并使用区域信息构造系数矩阵,使模型具有全局性。对MR(magneticresonance)图像分割的实验表明,该模型可以得到较好的分割结果,同时时间效率也有大幅提高,方便了实时应用。

The geodesic active contour model has been wide used in image segmentation for its computational stability.Most implicit active contour models,however,are based on explicit updating schemes and therefore of limited computational efficiency.To alleviate the problem,this paper presents a fast active contour model based on the additive operator splitting(AOS) algorithm.In the new model,the time step can be set large to improve the model’s time efficiency.During constructing the coefficient matrix,regional information is used to reduce the effect of noises.The results show that the new model can get better results in an efficient way.