引入邻域信息熵并用核距离代替简单的欧式距离,提出了邻域信息熵的核模糊C均值聚类图像分割算法(EKWFLICM).实验表明,该算法具有比WFLICM算法更强的去噪声能力,以使保留原图像细节的能力,具有一定实际应用价值.
This paper presents a new kernel fuzzy c-means clustering image segmentation algorithm withlocal entroy information (EKWFLICM)by introducing the pixel difference between neighborhood points and the middle point and kernel distance. The experimental results show that the proposed algorithm has betterr obustness than the WFLICM algorithm and retains the original image details well. So it has practical application value.