Volume 40 Issue 5
May  2015
Turn off MathJax
Article Contents
Abstract
References
Related Articles
ZHENG Zhaobao, PAN Li, ZHENG Hong. Application of Image Correlation Degree to Image Fuzzy Classification[J]. Geomatics and Information Science of Wuhan University, 2015, 40(5): 574-577. DOI: 10.13203/j.whugis20140736
Citation: ZHENG Zhaobao, PAN Li, ZHENG Hong. Application of Image Correlation Degree to Image Fuzzy Classification[J]. Geomatics and Information Science of Wuhan University, 2015, 40(5): 574-577. DOI: 10.13203/j.whugis20140736
shu

Application of Image Correlation Degree to Image Fuzzy Classification

  • 1School of Remote Sensing and Information Engineering,Wuhan University,Wuhan 430079,China;2Electronic Information School,Wuhan University,Wuhan 430072,China

Funds: The National 973Program of China,No.2012CB719905.
More Information
  • Author Bio:

    ZHENG Zhaobao: 国家973计划资助项目(2012CB719905) 或其他纹理特征的均值;然后,根据特征均值的大小将图像分成几类,假定分成3类(即有3类不同属性的图像),每类图像数量假定为n1、n2、n3计算每类图像的特征均值,接着计算每幅图像相对3个类别的关联度,所谓关联度,它反映一幅图像与所在类别和其他两个类别关系密切的程度在这个基础上,分别计算每个类别中所有图像关联度的均值和标准差σ,计算每幅图像的关联度与该类图像关联度均值之差Δi若Δi>2σ,则该图像“离开”该类别;否则,“保留”在该类别中“离开”图像应进入哪一类别由该图像相对另两类的关联度决定,哪一类的关联度大就进入那一类每幅图像都要经过“保留”与“离开”的检查,这样的检查意味着图像再分类1.2 图像与图像类别的关联度 假定有3组图像1、2、3组成一个组合,每组图像的像幅数分别为n1、n2、n3,每组图像纹理特征均值分别为mean1、mean2、mean3d1(i1,1)= mean1-Rbs(i1,1)d1(i1,2)= mean2-Rbs(i1,1)d1(i1,3)= mean3-Rbs(i1,1)(1) 计算每幅图像纹理特征相对3组图像纹理特征均值之差,即第一组图像特征相对第一、二、三组特征均值的差为d1(i1,1)、d1(i1,2)、d1(i1,3):式中,Rbs(i1,1)表示第一组图像中第i1 幅图像

  • Received Date: October 07, 2014
  • Revised Date: May 04, 2015
  • Published Date: May 04, 2015
    Graphical Abstract
    • Abstract
  • We propose a new image fuzzy classification method which based on the image relational de-gree.The correlation degree basis points are obtained in each category,and the mean and standarddeviationσof relational degree are calculated for each category.Then the correlation degree for eachimage and the mean difference of correlation degree for each categoryΔi are calculated.IfΔi≤2σ,theimage should“stay”in the class,otherwise,the image should be associated with a greater correlationdegree of the other two categories.Each image should have such inspection to achieve image re-divi-sion until the process is stable.Comparison results show that the quality of the quality of the imageclassification method has certain advantages.
    • FullText(HTML)
    • References (1)
  • [1]
    You J,Cohen H A.Classification and Segmentationof Rotated and Scaled Textured Images Using Tex-ture “Tuned”Mask[J].Pattern Recognition,1993,26(2):245-258[2] Zheng Zhaobao,Zheng Hong.Genetic Algorithmfor Producing Texture“Tuned”Masks[J].PR &AI,2001,14(1):119-122(郑肇葆,郑宏.产生纹理 “Tuned”模板的遗传算法[J].模式识别与人工智能,2001,14(1):119-122)[3] Zheng Zhaobao,Pan Li,Yu Xin.Bayesian Net-works for Producing Texture" Tuned" Masks[J].Geomatics and Information Science of Wuhan Uni-versity,2006,31(4):304-307(郑肇葆,潘励,虞欣.产生 “Tuned”模板的Bayesian Networks方法[J].武汉大学学报·信息科学版,2006,31(4):304-307)[4] Zheng Zhaobao.Honey-Bee Mating OptimizationAlgorithm for Producing Better“Tuned”Masks[J].Geomatics and Information Science of Wuhan Uni-versity,2009,34(4):387-390(郑肇葆.产生最佳 “Tuned”模板的蜜蜂交配算法[J].武汉大学学报· 信息科学版,2009,34(4):387-390)[5] Cordón O.A Proposal on Reasoning Method inFuzzy Rule-Based Classification System[J].Inter-national Journal of Approximate Reasoning,2009,20:21-45[6] Sanz J A.IVTURS:A Linguistic Fuzzy Rule-BasedClassification System Based on a New Interval-Val-ued Fuzzy Reasoning Method With Tuning and RuleSelection[J].IEEE Transactions on Fuzzy Sys-tems,2013,21(3):399-411[7] Zheng Zhaobao,Huang Guilan.Using Least SquareMethod for Texture Classification of Aerial Imageand Analysing Some Relative Problems[J].ActaGeodaetica tt Cartographica Sinica,1996,25(2):21-26(郑肇葆,黄桂兰.航空影像纹理分类的最小二乘法和问题分析[J].测绘学报,1996,25(2):21-26)
    • Related Articles

  • Related Articles

    [1]WANG Chunyan, LIU Jiaxin, XU Aigong, WANG Yu, SUI Xin. A New Method of Fuzzy Supervised Classification of High Resolution Remote Sensing Image[J]. Geomatics and Information Science of Wuhan University, 2018, 43(6): 922-929. DOI: 10.13203/j.whugis20150726
    [2]ZHENG Zhaobao, PAN Li, ZHENG Hong. Application of Image Correlation Degree to Image Fuzzy Classification[J]. Geomatics and Information Science of Wuhan University, 2015, 40(5): 574-577. DOI: 10.13203/j.whugis20140736
    [3]ZHENG Zhaobao, PAN Li, ZHENG Hong. Image Segmentation Based on Fuzzy Logic Methods[J]. Geomatics and Information Science of Wuhan University, 2014, 39(4): 397-400. DOI: 10.13203/j.whugis20120209
    [4]XU Lamei, YI Chunju. Application of Grey System Theory in Image Edge Detection[J]. Geomatics and Information Science of Wuhan University, 2012, 37(8): 929-931.
    [5]ZHONG Yanfei, ZHANG Liangpei, LI Pingxiang. Fuzzy Cluster Validation for Remote Sensing Image Classification[J]. Geomatics and Information Science of Wuhan University, 2009, 34(4): 391-394.
    [6]HU Peng, FU Zhongliang, CHEN Nan. Image Edge Detection Based on Gray System Theory[J]. Geomatics and Information Science of Wuhan University, 2006, 31(5): 411-414.
    [7]HE Jianhua, LIU Yaolin. Formalized Description of Indeterminate Geographic Object Based on Fuzzy Degree[J]. Geomatics and Information Science of Wuhan University, 2005, 30(4): 341-344.
    [8]ZHANG Huimin, HAN Liqun, DUAN Zhenggang. Tobacco Grading Based on Image Features[J]. Geomatics and Information Science of Wuhan University, 2003, 28(3): 359-362.
    [9]Huang Guilan, Zheng Zhaobao. The Application of Fuzzy Asesmble Analysis in the Classification of Image Texture Based on Fractal Geometry[J]. Geomatics and Information Science of Wuhan University, 1995, 20(2): 112-117.
    [10]Shen Bangxing. Fuzzy Comprehensive Evaluation of the Experimental Design for the Multi-target[J]. Geomatics and Information Science of Wuhan University, 1989, 14(1): 57-65.

  • Cited by

    Periodical cited type(1)

    1. 郑肇葆, 郑宏. 利用数据引力进行图像分类. 武汉大学学报(信息科学版). 2017(11): 1604-1607 .

    Other cited types(0)

Catalog

    Get Citation
    PDF
    XML
    Article views (4062) PDF downloads (1052) Cited by(1)
    Related

    Copyright © 2013 《武汉大学学报·信息科学版》编辑部

    Address：武汉大学文理学部本科生院楼北5楼Post Code：430072

    Tel：027-68778465,68788631E-mail:whuxxb@vip.163.com

    Submit Article

    Submission Guidelines

    e

    Contact Us

    Qrcode

    Supported by: Beijing Renhe Information Technology Co., Ltd.  Baidu Analytics

    /

    DownLoad:  Full-Size Img  PowerPoint
    Return
    Return