基于神经网络的交互式图像检索  

Interactive method of image retrieval based on neural networks

作　　者：王兵[1] 张欣[2] 王硕[1] 王苗[1] 

机构地区：[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

出　　处：《计算机工程与应用》2007年第22期59-62,共4页Computer Engineering and Applications

基　　金：河北省科技公关计划(the Key Technologies R&D Program of Hebei Province;China under Grant No.072135139);河北省教育厅项目 (No.Z2006414);河北省省级精品课建设项目

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。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 endusers 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.

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 

分 类 号：TP391.41[自动化与计算机技术—计算机应用技术]