Abstract: Color histograms are widely used for content-based image retrieval.Their advantages are efficiency, and insensitivity to small changes in camera viewpoint. However, a histogram is a coarse characterization of an image, and so images with very different appearances can have similar histograms.The paper describes a technique for comparing images called histogram refinement, which imposes additional constraints on histogram based matching. All the steps help to improve the image retrieval effect

Key words: Histogram; CBIR; Coherence vector; Method of clustering; Color space

摘要： 颜色直方图被广泛地应用在基于内容的图像检索中，其优点是效率高，对于视点细微的变化不敏感。然而直方图只是图像的一个粗糙特征，对于视觉上完全不相似的图像，其直方图有可能非常相似。该文在颜色直方图的基础上进行了改进，采用一种基于颜色聚合向量的图像检索方法，极大地提高了检索精度。

关键词: 直方图；基于内容的图像检索；聚合向量；聚类算法；颜色空间