本文提出了一种在获取图像低层视觉特征的基础上，利用语义网络对

图像进行语义自动分类，并在此基础上引入相关反馈技术，使图像的低级

物理特征和高级语义特征联系起来的算法，通过人一机协同工作，来弥补

计算机理解能力的不足，不断提高检索效果。

-In this paper, a low-level access to images based on visual characteristics, the use of semantic networks semantic image automatic classification, and on this basis the introduction of relevance feedback techniques, so that low-level image features and physical characteristics of high-level semantic link algorithm, through a machine to work together to make up for the lack of computer ability, continuously improve the search results.