首页 | 官方网站   微博 | 高级检索  
     

基于分块直方图和共生矩阵的图像检索方法
引用本文:李伟芳,滕奇志,汪华章.基于分块直方图和共生矩阵的图像检索方法[J].电子测量技术,2009(1):1-4.
作者姓名:李伟芳  滕奇志  汪华章
作者单位:四川大学电子信息学院图像信息研究所,成都,610064
摘    要:提出了一种基于分块直方图和共生矩阵相结合的图像检索方法。传统的基于颜色直方图特征的图像检索方法简单且易于实现,但是失去了颜色的空间分布信息。分块直方图法根据图像目标和背景的不同,将图像分成意义不同的8个子块,充分利用了图像分块的特点,改善了颜色特征缺乏空间信息的缺点。共生矩阵法根据不同图像间共生矩阵的差别来实现图像检索。最后,综合利用上述两个特征对图像进行共同检索。实验表明:综合2种特征检索图像比单一特征检索效果更好。

关 键 词:基于内容的图像检索  分块直方图  共生矩阵

Image retrieval based on blocked histograms and co-occurrence matrix
Li Weifang,Teng Qizhi,Wang Huazhang.Image retrieval based on blocked histograms and co-occurrence matrix[J].Electronic Measurement Technology,2009(1):1-4.
Authors:Li Weifang  Teng Qizhi  Wang Huazhang
Affiliation:Image Information Institute;Department of electronics and information;Sichuan University;Chengdu 610064
Abstract:A new method for image retrieval which was based on both blocked histograms and co-occurrence matrix was proposed.It was easy to perform for traditional image retrieval based on color histogram,however,its space distribution information was lost.Blocked histograms method was a feature extraction method based on the differences between the object and background,which divided the image into 8 different sub-blocks.This method amended the disadvantage of lacking spatial information of common color feature.The c...
Keywords:Content Based Image Retrieval (CBIR)  blocked histograms  matrix
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-23

京公网安备 11010802026262号