一种基于图像滤波的加权FCM图像分割算法

A kind of weighted FCM image segmentation algorithm based on image filtering

针对模糊C均值(FCM)图像分割算法受初始值影响较大以及对噪声的抑制作用较差的问题,提出一种基于图像滤波的加权FCM图像分割算法.该算法采用快速FCM算法进行初分割,降低了初始值的影响,同时引入自适应中值滤波器,并与加权FCM算法相结合进行迭代滤波分割,不仅能很好地抑制噪声的影响而且能使分割更精确.利用该算法分别对人工合成的和真实的含噪图像进行分割实验,实验结果表明:本文算法对含噪图像有很好的分割结果.

In view of the fuzzy c-means （ FCM） image segmentation algorithm is greatly influenced by the initial value of the poor and the inhibitory effect of noise, put forward a kind of weighted FCM image segmentation algorithm based on image filtering.At the beginning of the algorithm with fast FCM algorithm segmentation, to reduce the influence of the initial value, at the same time the adaptive median filter is introduced, and combined with weighted FCM algorithm segmentation, iterative filtering the effects of not only well restrain noise but also can make the segmentation more accurate.Respectively using the algorithm with noise of synthetic and real image segmentation experiments, the experimental results show that this algorithm for noise image has good segmentation results.