计算机工程与应用 ›› 2009, Vol. 45 ›› Issue (26): 147-149.DOI: 10.3778/j.issn.1002-8331.2009.26.043

• 图形、图像、模式识别 • 上一篇    下一篇

图像纹理联合关联规则挖掘研究

王佐成1,2,张飞舟1,薛丽霞2   

  1. 1.北京大学 遥感与地理信息系统研究所,北京 100871
    2.重庆邮电大学 软件学院,重庆 400065
  • 收稿日期:2009-03-18 修回日期:2009-04-20 出版日期:2009-09-11 发布日期:2009-09-11
  • 通讯作者: 王佐成

Research on image combined texture association rules mining

WANG Zuo-cheng1,2,ZHANG Fei-zhou1,XUE Li-xia2   

  1. 1.Institute of Remote Sensing and Geographic Information System,Peking University,Beijing 100871,China
    2.Software Institute,Chongqing University of Posts and Telecommunications,Chongqing 400065,China
  • Received:2009-03-18 Revised:2009-04-20 Online:2009-09-11 Published:2009-09-11
  • Contact: WANG Zuo-cheng

摘要: 提出以纹理联合关联规则来表达图像纹理特征以及挖掘纹理联合关联规则的算法。在纹理关联规则定义基础上,通过图像降噪预处理和数据挖掘预处理,采用模板统计挖掘方法挖掘低维和高维图像纹理联合关联规则。实验表明联合关联规则能够较好表达图像纹理特征,可以据此进行纹理分割。

关键词: 图像纹理, 数据挖掘, 关联规则, 联合关联规则

Abstract: The representation method of texture features by image combined texture association rules and the algorithm of mining image combined texture association rules are proposed.Based on the concept of image combined texture association rule,by noise reduction and data mining preprocessing,multi-dimension image combined texture association rules data mining based on mask of texture image counting can be accomplished.Experiments testify that image combined texture association rules can represent the texture features perfectly and texture image segmentation can be achieved.

Key words: image texture, data mining, association rule, combined texture association rules

中图分类号: