基于自适应相似度距离的改进FCM图像分割

Improved FCM image segmentation based on adaptive distance of similarity

为了改善传统FCM算法抗噪性差的问题,提出了基于自适应相似度距离的FCM算法。算法将像素分为两个特征:第一个描述的是像素的内在属性(灰度级特征),第二个描述邻域像素特征(空间特征)。在此基础上,基于自适应相似度距离,根据像素在图像中的空间位置决定哪一个特征拥有优先级,对其进行聚类。图像分割结果表明,算法比标准FCM算法有明显改善,具有很好的抗噪性能,取得了更好的分割效果。

Aiming at the poor noise resistance of FCM when applying to image segmentation, a fuzzy c-means clustering algo- rithm based on adaptive distance of similarity is proposed in this paper. The approach is proceeded by the characterization of pixels by two features： the first one describes the intrinsic properties of the pixel （spatial feature） and the second characterizes the neighborhood of pixel（ Grey level feature）. Then, the classification is made on the base on adaptive distance which privi- leges the one or the other features according to the spatial position of the pixel in the image. The obtained results have shown a significant improvement of performance compared to the standard version of FCM. Good anti-noise performance and better image segmentation effect are shown in the proposed algorithm.