基于多目标区间值模糊聚类的含噪图像分割  

A multi-objective interval valued fuzzy clustering algorithm for noisy image segmentation

作　　者：赵凤 李超琦 ZHAO Feng;LI Chaoqi(Key Laboratory of Electronic Information Application Technology for Scene Investigation,Ministry of Public Security,Xi an 710121,China;International Joint Research Center for Wireless Communication and Information Processing,Shaanxi,Xi an 710121,China;School of Communication and Information Engineering,Xi an University of Posts and Telecommunications,Xi an 710121,China)

机构地区：[1]电子信息现场勘验应用技术公安部重点实验室,陕西西安710121 [2]陕西省无线通信与信息处理技术国际合作研究中心,陕西西安710121 [3]西安邮电大学通信与信息工程学院,陕西西安710121

出　　处：《西安邮电大学学报》2018年第5期9-17,共9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;61671377;61102095);西安邮电大学新星团队资助项目(xyt2016-01)

摘　　要：针对区间值模糊聚类算法,提出基于多目标区间值模糊聚类的含噪图像分割算法。利用迭代非局部均值算法去除图像中的噪声点,并在其上构造一个区间值模糊图像,然后设计两个区间值模糊聚类有效性函数作为多目标优化的适应度函数,并构造一个融合区间值模糊信息的最优解选取指标。在自然图像和刑侦图像上的实验结果表明,该算法能够在抑制噪声的同时取得良好的分割效果。A noisy image segmentation algorithm based on the multi-objective interval valued fuzzy clustering is proposed.The iterative non-local means algorithm is used to remove the noise of noisy image and an interval valued fuzzy image is thus constructed.Then two interval valued fuzzy cluster validity functions are designed as fitness functions of the multi-objective optimization,and a cluster validity index using the interval valued fuzzy information is therefore constructed.Experiment results on real images and criminal investigation images show that the proposed method can obtain the better segmentation effect while suppressing the noise.

关 键 词：图像分割 多目标优化 区间值模糊聚类 非局部空间信息 

分 类 号：TP391.4[自动化与计算机技术—计算机应用技术]