【摘要】 提出了一种基于Rough集理论的语义图像检索(RSBSIR)方法.对于给定的图像集和一个已经分类的图像集,能够迅速区别图像类型的最小的并列关键词集,降低了关键词向量空间的维数,缩小了问题的规模,简化了基于知识库的图像检索系统的建立过程.并且对于用户而言,小的关键词集易于理解,同时对规则库也可以方便地进行编辑.更多还原

【Abstract】 A method for semantic image retrieval based on Rough set was presented.For given image sets and a categorized image set,the minimal keyword union sets of image classes were differentiated rapidly byusing this method.The di mension of keyword vectors and the scope of the problem were reduced by this method,also.The creation process of knowledge-base-based image retrieval system was reducted,too.For the users,these reduced keyword sets were easy to understand,the warehouse of the rules were easy toedit at the same time.更多还原