增强型区间二型FCM算法.pdfVIP

第29 卷 第3 期 控 制 与 决 策 2014 年 3 月 Vol. 29 No. 3 Control and Decision Mar. 2014 文章编号: 1001-0920 (2014) 03-0465-05 DOI: 10.13195/j.kzyjc.2012.1778 增强型区间二型FCM 算法 邱存勇, 肖 建, 韩 璐 (西南交通大学电气工程学院，成都610031) 摘 要: 不确定性存在于图像处理、模式识别等众多领域的实际应用中, 模糊 均值聚类(FCM) 算法虽广泛应用于 这些领域, 但其处理不确定性的能力较差. 引入区间二型模糊理论能有效提升算法处理不确定性的能力, 但相应地造 成算法复杂度增加, 制约了区间二型FCM 算法的推广应用. 鉴于此, 提出增强型区间二型FCM 算法, 通过优化初始 聚类中心和降型运算, 极大地减少了区间二型FCM 算法的运算量, 并提升算法的收敛速度. 通过对随机和实际数据 的实验比较验证了改进算法的有效性. 关键词: 模糊聚类；模糊 均值聚类；区间二型；降型 中图分类号: TP391.4 文献标志码: A Enhanced interval type-2 fuzzy -means algorithm QIU Cun-yong, XIAO Jian , HAN Lu (School of Electrical Engineering，Southwest Jiaotong University，Chengdu 610031，China ．Correspondent ：QIU Cun- yong ，E-mail ：chaucer qcy@my.swjtu.edu.cn) Abstract: There are various uncertainties in the applications in many fields such as image processing and pattern recognition. However, the fuzzy -means(FCM) algorithm which is widely used in these fields cannot handle the uncertainties well. The introduction of interval type-2 fuzzy theory into FCM can bring such algorithm the ability of handling uncertainties, but the complexity of algorithm will increase accordingly. In order to reduce the complexity, an enhanced interval type-2 FCM algorithm is proposed. The initialization of cluster center and the process of type-reduction are optimized in this algorithm, which can greatly reduce the calculation of interval type-2 FCM and accelerate the convergence of the algorithm. The experimental results on random data and real data show the effectiveness of our proposed algorithm. Key words: fuzzy cluster ；fuzzy -means ；interval typ