基于本体和奇异值分解的图像数据索引查询算法

摘要

An image data query algorithm based on latent semantic indexing in domain ontology and singular value decompo-sition is proposed .It projects query expansion vector on latent semantic area .A new method to computing similarity between query vectors and query documents is proposed .Users can get accurate and effective retrieval results using the similarity .The experimental results show that the proposed solution can improve image index query recall ratio and precision ratio .%提出一种基于领域本体潜在语义索引和奇异值分解的图像数据查询算法，将查询扩展向量映射到潜在语义空间，根据相似度计算方法计算查询向量与图像文档之间的相似度，并将相似度大于阀值的文档作为检索结果降序排列返回给用户。该算法能更有效地提高图像检索的查准率和查全率。