摘要： 对于当前基于内容的图象检索（ＣＢＩＲ）系统来说，在用户的信息需求和系统的功能之间存在着重大的差距。要进一步提高现有图象检索系统的能力，需要对图象的内容进行语义描述。对图象的多层语义模型和图象语义的知识表示作了概括性介绍，重点讨论了一些将图象的低层视觉特征映射到图象高层语义的方法，并指出了利用语义进行图象检索急需解决的一些问题。

关键词: 图象语义, 语义模型, 知识表示, 基于内容的图象检索

Abstract: In image retrieval system today, there exists a significant gap between the user's information needs and what the systems can deliver. To increase the image retrieval system's ability, we require a proper semantic description for image content. In this article, we give a comprehensive discuss about multi-level semantic modeling and knowledge representation in image retrieval, especially about the methods with which to mapping visual features to high lever image semantic contents. To make image retrieval using semantics efficiently and effectively, we point out the research progress we need to make at present and future.

中图分类号: 

• TP391.3