All Title Author
Keywords Abstract


基于SSCL的模糊C均值图像分类方法

DOI: 10.11834/jig.20110201

Keywords: 图像分割,模糊C均值,自分裂竞争学习

Full-Text   Cite this paper   Add to My Lib

Abstract:

针对传统模糊C均值聚类方法对噪声敏感和过分依赖于初始聚类中心的缺点,提出基于SSCL的模糊C均值图像分类的自适应算法。该算法首先通过SSCL获得初始类别数和类别中心,然后作为模糊C均值聚类的输入,自动对图像进行分割,并对图像分割结果利用空间信息进行后处理。实验结果表明该方法较好地解决了FCM算法中的初始化和噪声敏感问题,具有较好的分类结果。

References

[1]  Lin, TC,Partition belief median filter based on Dempster-Shafer theory for image processing,Pattern Recognition?,2008, 41(1).
[2]  Fayed, HA ;Hashem, SR ;Atiya, AF,Self-generating prototypes for pattern classification,Pattern Recognition?,2007, 40(5).
[3]  Bezdek J C,Pattern Recognition with Fuzzy Objective Function Algorithms,New York:Plenum,1981.
[4]  Lu Yingli;Jiang Tanzi;Zang Yufeng,A split-merge based regiongrowing method for mri activation detection,Human Brain Mapping,2004(31).
[5]  Masson, MH ;Denoeux, T,ECM: An evidential version of the fuzzy c-means algorithm,Pattern Recognition?,2008, 41(4).
[6]  Zhang Yajun;Liu Zhiqiang,Self-splitting competitive learning:a new on-line clustering paradigm,IEEE Transactions on Neural Networks,2002(02).
[7]  Richard O D;Peter E H;DavidG S,模式分类,北京:机械工业出版社,2003.
[8]  Dunn J C,A fuzzy relative of the isodata process and its use in detecting compact well-separated clusters,J Cybem Infsci (S0146-5090),1974(03).
[9]  郝颖明;朱枫.2维Otsu自适应阈值的快速算法[J].中国图象图形学报,2005(04)

Full-Text

comments powered by Disqus