• 中国计算机学会会刊
  • 中国科技核心期刊
  • 中文核心期刊

J4 ›› 2007, Vol. 29 ›› Issue (4): 55-58.

• 论文 • 上一篇    下一篇

基于HSI色彩空间累加直方图的图像检索算法

王剑峰 肖国强 江健民   

  • 出版日期:2007-04-01 发布日期:2010-05-30

  • Online:2007-04-01 Published:2010-05-30

摘要:

本文给出了一种基于HSI色彩空间累加直方图进行图像检索的方法。该方法使用符合人类视觉特性的HSI颜色模型将颜色空间压缩为72种具有代表性的颜色,设计了一种具有重叠方式的图像分块方法;然后统计每块区域内的颜色直方图得到该图像的HSI颜色空间直方图,再计算各块的累加直方图,并根据相似度计算方法计算每块子图像的累加直方图和图像库中图像对应块的累加直方图的相似度;最后根据各块子图像的重要性进行加权,得出最后的检索结果。实验表明,该方法实现简单,能够更加灵活、准确地描述图像的颜色特征,从而有效提高了检索的精度。

关键词: 基于内容的图像检索 HSI色彩空间 累加直方图

Abstract:

This paper gives an image retrieval method based on the color-space histogram.First,it uses the HSI color space to  compres the color space to 72 kinds of representative colors,and then designs 12 overlapped sub-regions to get the color accumulative histogram in each region. Finally, the similarity iscalculated based on the accumulate.re histogram of each region. The proposed method is simple, flexible and effective in describing the color feature,  which avoids the shortcoming of losing the space distribution in formation in the color histogram. The experimental resuhs show the superiority of the p roposed algorithm in image retrieval precision.

Key words: content-based image retrieval (CBIR);HSI color space;accumulative histogram