摘要
在基于内容的图像检索系统中我们经常使用一些底层特征,如表现图像的颜色和文本信息。如果给出了这些特征的内容,如特征向量,我们就可以通过计算特征空间的距离测量出图像之间的相似度。然而,这些底层特征并不一定能反映人的视觉中高层概念的相似度。相关反馈技术就是通过交互检索来提高检索性能,在数据库的搜索中参考了用户的反馈信息。因此提出了一种加权距离法,数据库中的图像与用户选择的那些相关图像的特征值的标准偏差的比率就是该图像的加权值。反馈技术不仅适用于相互独立的权值还适用于增量更新的权值,并且这些权值反过来也改善了不同特征在数据库检索中的效果。实验中用平均查准率和增进来评估检索性能,在有1000幅图像的数据库中,首轮交互之后检索的性能就平均提高了19%。
Content-based image retrieval systems use low-level features like color and texture for image representation.Given these representations as feature vectors,similarity between images is measured by computing distances in the feature space.However,these low-level features cannot always express the high-level concept of similarity in human perception.Relevance feedback tries to improve the performance by introducing iterative retrievals where the feedback information from the user is incorporated into the database search.We present a weighted distance approach,where the weights are the rations of standard deviations of the feature values both for the whole database and also among the relevant images selected by the user.The feedback is used for both independent and incremental updating of the weights and these weights are used to iteratively refine the effects of different features in the database search.Retrieval performance is evaluated using average precision and progress that are computed on a database of 1,000 images and an average performance improvement of 19% is obtained after the first iteration.
出处
《电子测试》
2011年第5期40-43,共4页
Electronic Test
关键词
相关反馈
加权距离
图像检索
relevance feedback
weighted distance
image retrieval
