图像检索中的关键技术

Key-Techniques in Content-Based Image Retrieval

传统的基于数值 字符的信息检索技术并不能满足海量的图像检索要求 ,因此 ,基于内容的图像检索技术 (CBIR ,content basedImageretrieval)得到了广泛研究。本文主要讨论CBIR研究中的一些关键问题 :图像的内容特征及其提取、特征之间的相似度计算、查询条件的表达等 ,最后指出了一些可值得深入研究的方向。

Because of the complexity of images,the traditional information retrieval techniques,which are based on value and/or text comparison, are not sufficient to retrieval images,the research is shift to the techniques of content based image retrieval (CBIR).Some key issues on CBIR are discussed:image content extraction,image features similarity measures,the expression of query conditions,and so on.Some future research directions are also given in the conclusion.