基于关联规则挖掘的图像检索

针对图像检索中图像相似性判断所面临的"语义鸿沟"问题,本文提出了基于关联规则挖掘的图像检索。该方法利用数据挖掘技术,对图像纹理的各种低层可视化特征进行分析挖掘,获得一系列的图像低层特征值与高层概念之间的各种规则,将这些规则构成图像知识库,并与概念集合相对应。实验结果表明,采用关联规则挖掘能很好的提高图像检索的准确率。

Image Retrieval Based on Association Rule Mining

For image retrieval,image similarity judgments facing the "semantic gap" problem,this paper presents the image retrieval based on association rule mining.The method uses data mining techniques,a variety of low-level visual features of image texture analysis and mining,access to a variety of rules in a series of low-level image characteristic value and high-level concept,these rules constitute the image knowledge base,and corresponding to the concept collection.the experimental results show that association rule mining to improve the accuracy of image retrieval.