基于局域模糊聚类和Chan-Vese模型的CT医学图像分割

针对CT医学图像灰度不均匀的特点,研究了基于改进的模糊聚类和ChanVese模型的图像分割.该分割模型综合利用基于空间信息的FCM算法、图像局部区域信息以及Chan-Vese模型,通过最小化能量函数的方式来进行曲线演化.基于空间信息的FCM算法对曲线的演化起到了一定的收敛作用,并且局部区域信息提高了分割质量.分割模型还考虑了分割效果和计算效率,降低了算法的时间复杂度,提高了算法的执行效率,从而提高了灰度不均匀图像分割的精度.

A Local Fuzzy-based Chan-Vese Method for the Segmentation of CT Medical Images

Due to the inhomogeneous density of CT medical image,the segmentation model based on the local fuzzy clustering method and Chan-Vese model is investigated.This segmentation model makes full use of the spatial fuzzy c-means (SFCM) algorithm,the local region information,and local Chan-Vese model.The curve evolution is realized by minimizing the energy function.SFCM algorithm is convenient for controlling level set evolution and calculation convergence,and the local image information can improve the quality of segmentation.Moreover,the segmentation model considers the effectiveness of segmentation and high efficiency to reduce the time complexity of the algorithm.Thus,the segmentation model gets more precise result of image segmentation for the inhomogeneous density of CT medical image.