计算机工程与应用 ›› 2010, Vol. 46 ›› Issue (21): 165-167.DOI: 10.3778/j.issn.1002-8331.2010.21.047

• 图形、图像、模式识别 • 上一篇    下一篇

采用位平面直方图特征向量的图像检索方法

杨红菊,韩建栋,曹付元   

  1. 山西大学 计算机与信息技术学院 计算智能与中文信息处理教育部重点实验室,太原 030006
  • 收稿日期:2010-01-12 修回日期:2010-05-24 出版日期:2010-07-21 发布日期:2010-07-21
  • 通讯作者: 杨红菊

Method of image retrieval based on bit-plane histogram features vector

YANG Hong-ju,HAN Jian-dong,CAO Fu-yuan   

  1. Key Laboratory of Ministry of Education for Computational Intelligence and Chinese Information Processing,School of Computer and Information Technology,Shanxi University,Taiyuan 030006,China
  • Received:2010-01-12 Revised:2010-05-24 Online:2010-07-21 Published:2010-07-21
  • Contact: YANG Hong-ju

摘要: 首先将图像分解为8个位平面,选择前4个重要位平面,求出其灰度码表示,根据每个灰度码位平面的颜色直方图,计算均值、标准偏差、偏斜度、能量、熵;综合这些特征构成名为位平面直方图特征向量的组合特征,进行图像检索。实验中采用Tonimoto相似度量函数计算图像间的相似度。该方法计算速度快,避免了图像量化造成的误检。实验结果显示了该方法的检索性能。

关键词: 图像检索, 位平面, 位平面直方图特征向量, 直方图特征, 灰度码

Abstract: Firstly,the image is divided into eight bit planes.The first four most significant bit planes of the image are considered and later each 4-bit vector is converted to its corresponding reflected gray code.Histogram features of each bit plane are computed:Mean,standard deviation,skew,energy and entropy.Then,a combined vector named bit-plane histogram features vector is constructed for image retrieval.Finally,Tonimoto measure is adopted to measure the similarity between images.The method is fast and has no improper retrieval caused by image color quantization technique.Experimental results demonstrate the performance of the method.

Key words: image retrieval, bit-plane, bit-plane histogram features vector, histogram features, gray code

中图分类号: