计算机工程与应用 ›› 2007, Vol. 43 ›› Issue (22): 59-62.

• 学术探讨 • 上一篇    下一篇

基于神经网络的交互式图像检索

王 兵1,张 欣2,王 硕1,王 苗1   

  1. 1.河北大学 数学与计算机学院,河北 保定 071002
    2.河北大学 电子信息工程学院,河北 保定 071002
  • 收稿日期:1900-01-01 修回日期:1900-01-01 出版日期:2007-08-01 发布日期:2007-08-01
  • 通讯作者: 王 兵

Interactive method of image retrieval based on neural networks

WANG Bing1,ZHANG Xin2,WANG Shuo1,WANG Miao1   

  1. 1.College of Mathematics and Computer Science,Hebei University,Baoding,Hebei 071002,China
    2.College of Electronics and Information Engineering,Hebei University,Baoding,Hebei 071002,China
  • Received:1900-01-01 Revised:1900-01-01 Online:2007-08-01 Published:2007-08-01
  • Contact: WANG Bing

摘要: 提出了基于神经网络的交互式图像检索方法,系统根据用户对检索结果的评价,动态构造神经网络,描述图像之间的相似性;图像间的这种相似性以及本次检索结果可以作为以后检索的历史信息保存在神经网络中,从而提高下一次检索的效率。实验表明,该方法嵌入到典型的图像检索系统中,改善了图像检索性能。

关键词: 神经网络, 基于内容的图像检索, 相关反馈, 集合的划分

Abstract: An interactive image retrieval based on neural networks is employed to measure the similarities between the query image and the images in a database.The neural networks presented in this paper are constructed dynamically according the end-users relevance feedbacks.These similarities and historical relevance feedbacks deposited in the neural network can be used in next retrievals to gain enhancement in the efficiency of the system retrieval.This approach may be embedded in many current image retrieval systems to improve the performance of the retrieval.Results in our experimentation demonstrate a good potential of the methodology.

Key words: neural networks, content-based image retrieval, relevance feedback, set partition