首页 | 本学科首页   官方微博 | 高级检索  
     

基于统计特征的DCT压缩域纹理图像检索方法
引用本文:张曦煌,卞国春,李红.基于统计特征的DCT压缩域纹理图像检索方法[J].计算机工程与设计,2006,27(7):1282-1285.
作者姓名:张曦煌  卞国春  李红
作者单位:江南大学,信息工程学院,江苏,无锡,214122
摘    要:提出了一种基于离散余弦变换(Discrete Cosine Transfrom,DCT)的纹理图像的检索方法.该方法在DCT压缩域,通过直接对DCT系数计算,获得图像纹理的统计特征,并作为检索的依据.理论分析和实验结果都表明,该方法具有很好的检索准确率和效率,并且对于旋转具有不变性.

关 键 词:DCT变换  压缩域  纹理  统计特征  图像检索
文章编号:1000-7024(2006)07-1282-04
收稿时间:2005-01-16
修稿时间:2005-01-16

Algorithm of texture images retrieval in DCT compressed domain based on statistical features
ZHANG Xi-huang,BIAN Guo-chun,LI Hong.Algorithm of texture images retrieval in DCT compressed domain based on statistical features[J].Computer Engineering and Design,2006,27(7):1282-1285.
Authors:ZHANG Xi-huang  BIAN Guo-chun  LI Hong
Affiliation:School of Information Engineering, Southern Yangtze University, Wuxi 214122, China
Abstract:A new algorithm of retrieval for texture images in DCT compressed domain is presented. The statistical features of the texture images is computed directly from DCT coefficients, and used for image retrieval. Theoretical analysis and experimental results show that the approach achieves good performances in terms of retrieval accuracy and effectiveness. It is also invariant to rotation.
Keywords:DCT  compressed domain  texture  statistical feature  image retrieval
本文献已被 CNKI 维普 万方数据 等数据库收录!
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