首页 | 本学科首页   官方微博 | 高级检索  
     

基于图像能量谱直方图的纹理检索算法RAH
引用本文:丁玲,王崇俊,杨育彬,陈世福.基于图像能量谱直方图的纹理检索算法RAH[J].计算机科学,2005,32(1):194-197.
作者姓名:丁玲  王崇俊  杨育彬  陈世福
作者单位:南京大学计算机软件新技术国家重点实验室,南京,210093;南京大学计算机软件新技术国家重点实验室,南京,210093;南京大学计算机软件新技术国家重点实验室,南京,210093;南京大学计算机软件新技术国家重点实验室,南京,210093
基金项目:江苏省应用基础基金BJ20009
摘    要:本文提出的基于纹理的图像检索算法RAH是以图像的能量谱直方图为基础,包括以下两个方面:第一,计算图像能量谱的半径和角度直方图的方法以及基于图像能量直方图的频率去提取纹理特征的算法;第二,基于纹理特征的相似度检测方程。这个实验结果表明,RAH算法优于一般的灰度共生矩阵纹理检索算法,具有较好的检索效果,比较适用于基于内容的图像检索。

关 键 词:基于内容的图像检索  提取纹理特征  灰度共生矩阵  相似度检测方程

RAH: A Texture Based Image Retrieval Algorithm Based on the Histogram of Image Power Spectral
DING Ling,WANG Chong-Jun,YANG Yu-Bin,CHEN Shi-Fu State Key Laboratory for Novel Software Technology,Nanjing University,Nanjing.RAH: A Texture Based Image Retrieval Algorithm Based on the Histogram of Image Power Spectral[J].Computer Science,2005,32(1):194-197.
Authors:DING Ling  WANG Chong-Jun  YANG Yu-Bin  CHEN Shi-Fu State Key Laboratory for Novel Software Technology  Nanjing University  Nanjing
Affiliation:DING Ling,WANG Chong-Jun,YANG Yu-Bin,CHEN Shi-Fu State Key Laboratory for Novel Software Technology,Nanjing University,Nanjing 210093
Abstract:A texture based image retrieval algorithm RAH based on the histogram of image power spectral in frequen- cy domain is proposed, which mainly includes the next two parts: first, the method of computing radius histogram and angular histogram and the algorithm of texture feature extraction based on the histogram of image power spectral in frequency domain, second, the similarity measurement function based on the texture feature. The experimental re- sults indicate that RAH is preferable to the traditional co-occurrence matrix based algorithm.
Keywords:Content-based image retrieval  Texture feature extraction  Co-occurrence matrix  Similarity measurement function
本文献已被 CNKI 维普 万方数据 等数据库收录!
点击此处可从《计算机科学》下载全文
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