Abstract: A new image retrieval method based on region is proposed in this paper.First，a method of color clustering is employed to divide images into regions and low-level feature for the color，position，shape of each region are subsequently extracted.Then this paper implements similarity measurement of two images by taking into consideration the similarity of every region synthetically.In order to improve the precision of image retrieval，a relevance feedback mechanism，based on support vector machines is invoked.The experiment results show that the color image retrieval based on object regions is superior to that based on global image features.

Key words: image retrieval, image segmentation, relevance feedback, support vector machines

摘要： 提出一种基于目标区域的图像检索方法，首先采用颜色聚类的分割方法将图像分割成不同的区域，提取每个区域的颜色、位置、形状等低层特征，然后提出一种相似度计算方法实现图像的相似性度量。为了提高图像检索的准确度，最后采用支持向量机（SVM）的相关反馈算法。实验结果表明，基于目标区域的图像检索效果比基于全局图像特征的检索效果有较好的改善。

关键词: 图像检索, 图像分割, 相关反馈, 支持向量机