随着计算机科技的发展，图像检索的应用也越来越成熟，根据检索性质可分为两类：基于文本的图像检索和基于内容的图像检索。本论文通过研究基于内容的图像检索中的几个核心算法，用于聚类分析的K-means算法，通过haar小波变换来提取图像底层视觉特征，以及使用F-范数理论来进行相似性度量，来设计一个离线的图像检索系统。(With the development of computer technology, the application of image retrieval is more and more mature. According to the nature of retrieval, they can be divided into two categories: text based image retrieval and content based image retrieval. This paper studies the core algorithm in content based image retrieval, K-means algorithm for clustering analysis, through Haar wavelet transform to extract image visual features, and using F- norm theory to similarity measure, image retrieval system to design an offline.)