互信息在图像检索中的应用

Application of Mutual Information in Image Retrieval

介绍了一种用互信息来衡量相似性图像检索方法。该方法首先生成一种在统计上有代表性的视觉模式,使用这种模式的分布作为图像内容的描述符;基于该内容描述,设计了其互信息的计算方法以衡量图像的相似性。实验结果表明,在图像检索中,相对于其它如KL散度和L2规范等方法,互信息是一种更为有效的衡量相似性的方法。

An approach for image retrieval using mutual information as a similarity measure is presented in this paper.It is based on the premise that two similar images should have high mutual information,or equivalently,the querying image should convey high information about those similar to it.The method first generates a set of statistically representative visual patterns and uses the distributions of these patterns as images content descriptors.To measure the similarity of two images,we develop a method to compute the mutual information between their content descriptors.Two images with larger descriptor mutual information is regarded as more similarity.The experimental results demonstrate that mutual information is a more effective image similarity measurement than others such as Kullback-Leibler divergence and L2 norms.