针对图像检索中的低层视觉特征相似性度量问题，提出一种基于语义测度的图像相似性计算方法。该方法在图像区域分割的基础上，通过构建图像区域子块与语义元数据之间的统计映射关系，实现图像内容的统计语义描述，建立图像之间、图像与语义类别、语义类别之间的分层语义相似测度。通过对自然图像库的实验结果表明，该方法在相似图像检索中具有更好的性能。

In order to solve the problem that the low-level visual features can not express the high-level semantic contents accurately in image retrieval, the paper presents a new image similarity calculating approach using the proposed semantic measure. Based on the segmented image regions, the statistical mapping relationships between image blobs and semantic metadata are constructed and the layered semantic measurements between images are then defined. Experiment results on natural images show good retrieval quality based on the semantic similarity measure method.