期刊名称：International Journal of Innovative Research in Computer and Communication Engineering
印刷版ISSN：2320-9798
电子版ISSN：2320-9801
出版年度：2013
卷号：1
期号：5
出版社：S&S Publications
摘要：Typical content-based image retrieval (CBIR) system query results are a set of images sorted by featuresimilarities with respect to the query. However, images with high feature similarities to the query may be very differentfrom the query in terms of semantics. This is known as the semantic gap. We introduce a novel image retrieval schemeCLUster-based rEtrieval of images by unsupervised learning which tackles the semantic gap problem based on ahypothesis: semantically images tend to be clustered in some feature space. CLUE attempts to capture semanticconcepts by learning the way that images of the same semantics are similar and retrieving image clusters instead of aset of ordered images. Clustering in CLUE is dynamic. In particular, clusters formed depend on which images areretrieved in response to the query therefore; clusters give the algorithm as well as the user’s semantic relevant clues asto where to navigate. CLUE is a general approach that can be combined with any real-valued symmetric similaritymeasure. Thus it may be embedded in many current CBIR systems. Experimental results based on a database of about60,000 images from COREL demonstrate improved performance.
