To overcome the shortcomings of traditional text-based query,content-based retrieval is specially emphasized. In this paper, the architecture of image database is discussed; the past, the present and the future of image retrieval are analyzed, and the main issues at different research phases are also discussed. The key research issues of content-based image retrieval are discussed extensively and comprehensively.Now most "content-based image retrieval" techniques rely on global attributes such as color or texture histograms, which tend to ignore the spatial composition of the image. On the basis of researching image feature representation and index technology, the paper present an alternative image retrieval system based on the principle that it is the user who is most qualified to specify the query "content" . With new system, the user can define multiple "region of interest" and can specify the relevance of their spatial layout in the retrieval process. The methods often do succeed in capturing higher-level knowledge about what exactly was of interest to the user in the query image; consequently reduce the blindness of retrieval.
|