首页 | 本学科首页   官方微博 | 高级检索  
     检索      

遥感图像纹理关联规则挖掘
引用本文:王佐成,薛丽霞,凌聪.遥感图像纹理关联规则挖掘[J].重庆邮电大学学报(自然科学版),2007,19(4):438-441.
作者姓名:王佐成  薛丽霞  凌聪
作者单位:1. 重庆邮电大学,软件学院,重庆,400065
2. 重庆邮电大学,计算机科学与技术学院,重庆,400065
摘    要:基于遥感图像的光谱特征和纹理特征,提出图像关联规则的定义,通过对图像降阶预处理构造图像模板, 采用关联规则挖掘算法对图像纹理的频繁模式进行挖掘,通过联合关联规则来表达纹理。试验表明,挖掘出的关 联规则不仅能够表达规则纹理,而且能够较好表达随机纹理,据此可以进行纹理图像分割。

关 键 词:数据挖掘  纹理  图像  关联规则
文章编号:1673-825X(2007)04-0438-04
收稿时间:2006-10-09
修稿时间:2006-10-092007-06-19

Mining textural association rules in RS image
WANG Zuo-cheng,XUE Li-xi,LING Cong.Mining textural association rules in RS image[J].Journal of Chongqing University of Posts and Telecommunications,2007,19(4):438-441.
Authors:WANG Zuo-cheng  XUE Li-xi  LING Cong
Institution:1. School of Software, Chongqing University of Posts and Telecommunications, Chongqing 400065, P. R. China; 2.College of Computer Science and Technology, Chongqing University of Posts and Telecommunications, Chongqing 400065, P. R. China
Abstract:Based on the gray and texture feature of remote sensing image, this paper proposes the concept of association rule in image. By degrading the image, the mask of image can be constructed, and the frequent patterns of texture can be mined by the algorithm of association rules mining. The association rules can be combined to represent the texture. The experimental results validate that the combined association rules can represent the regular texture and represent the random texture perfectly. Therefore, the image segmentation can be accomplished by the combined association rules.
Keywords:data mining  texture  image  association rule
本文献已被 维普 万方数据 等数据库收录!
点击此处可从《重庆邮电大学学报(自然科学版)》浏览原始摘要信息
点击此处可从《重庆邮电大学学报(自然科学版)》下载免费的PDF全文
设为首页 | 免责声明 | 关于勤云 | 加入收藏

Copyright©北京勤云科技发展有限公司  京ICP备09084417号