基于智能体和本体的WEB图像检索模型  

WEB IMAGE RETRIEVAL MODEL BASED ON AGENT AND ONTOLOGY

作　　者：陈平[1] 李垒[1] 任越美[2] 程显毅[2] 

机构地区：[1]河南工业职业技术学院,河南南阳473009 [2]江苏大学,江苏镇江212013

出　　处：《计算机应用与软件》2009年第7期34-36,共3页Computer Applications and Software

基　　金：国家自然科学基金(60473039)

摘　　要：针对基于关键字WEB图像检索中的语义缺失问题,利用本体的方法描述WEB图像的语义特征,构建了基于智能体和语义特征的WEB图像检索模型,该模型以领域Ontology描述WEB图像的语义特征,通过多个Agent模块分工协作,完成满足用户请求的WEB图像检索。并在Corel提供的图像上进行了仿真实验,验证了该模型解决了基于关键字WEB图像检索模型中的语义缺失问题,提高了WEB图像检索速度和准确率。For remedying the semantics deficiency of the keywords-based WEB image retrieval, using ontology to describe the semantic features of WEB images, a new WEB image retrieval model based on agent and semantic features was constructed. This model describes semantic features of WEB images by domain Ontology ,through work division and coordination among multi-agent modules to complete the WEB image retrieval which satisfies users＇ requests. Furthermore, the simulation experiment on images provided by Corel was performed,it confirmed that this model has solved the semantic deficiency problem of keywords-based WEB image retrieval model and enhanced the speed and precision of WEB image retrieval.

关 键 词：本体 语义特征 智能体 图像检索 

分 类 号：TP391.41[自动化与计算机技术—计算机应用技术] G354[自动化与计算机技术—计算机科学与技术]