基于SPKFCM快速双循环水平集非均匀图像分割

针对传统快速双循环水平集对初始演化曲线过于依赖的问题,提出一种基于空间惩罚核模糊C-means (SPKFCM)算法的初始演化曲线自动选取快速双循环水平集算法.首先,对模糊均值聚类算法进行改进,通过增加空间惩罚函数提出SPKFCM算法,用于对快速双循环水平集算法的自动初始化;其次,基于SPKFCM并结合快速双循环水平集算法,设计基于SPKFCM快速双循环水平集算法框架,并给出相应速度参量Fd和Fint模糊化形式;最后,通过与已有算法在仿真图像上的对比结果显示,所提算法在随机初始化条件下,具有更高的分割精度和计算效率.

SPKFCM based fast two-cycle level set non-uniform image segmentation

According to the problem of the traditional fast two-cycle level set too dependent on the initial condition, here proposed a kind of space punishment core based c-means algorithm for the initial evolution curve automatically selection for fast two-cycle level set. Firstly, it took the improvement to the fuzzy c-means clustering algorithm and increasing the space penalty core function to propose SPKFCM algorithm, which was used for fast two-cycle level set automatic initialization; Secondly, combined SPKFCM and fast two-cycle level set algorithm, here designed the SPKFCM based fast two-cycle level set algorithm, and gave the fuzzy form of the corresponding speed parameters and ; Finally, through simulation with existing contrast algorithms showed that the proposed algorithm had higher degree of segmentation precision and higher computation efficiency.