摘要
针对抑制式模糊C-均值聚类算法的参数选择问题,采用一种直觉模糊补函数生成方法获得犹豫度,通过将其作为抑制因子实现参数的自适应选取,进而给出一种基于犹豫度生成的抑制式模糊C-均值图像分割算法。测试图像的实验结果显示,与模糊C-均值和一种直觉模糊C-均值聚类算法相比,该算法的迭代次数和运行时间均有一定下降,且能够更好地保持图像的细节信息。
For dealing with the parameters selection of suppressed FCM algorithm,a kind of fuzzy complementary function is used to generate hesitancy degree.By regarding the generated hesitation degree as suppression parameter,the parameter can be selected adaptively.Then,a suppressed FCM for image segmentation based on the generated hesitation is suggested.The experimental results of images show that the iteration number and running time of the proposed algorithm are less than those of the FCM and intuitionistic fuzzy sets-fuzzy c-means clustering algorithms,and the proposed algorithm can maintain image＇s detail information better.
出处
《西安邮电大学学报》
2017年第5期50-55,共6页
Journal of Xi’an University of Posts and Telecommunications
基金
国家自然科学基金资助项目(61571361)
陕西省教育厅科学研究计划资助项目(16JK1709)
西安邮电大学西邮新星团队资助
关键词
抑制式模糊C-均值聚类
直觉模糊集
抑制因子
犹豫度
suppressed fuzzy c-mean clustering,intuitionistic fuzzy sets,suppression parameter, hesitation degree
