A Scheme for Intelligent Image Retrieval in Multimedia Databases

✍ Scribed by Jesse S. Jin; Ruth Kurniawati; Guangyu Xu

• Publisher: Elsevier Science
• Year: 1996
• Tongue: English
• Weight: 978 KB
• Volume: 7
• Category: Article
• ISSN: 1047-3203
• DOI: 10.1006/jvci.1996.0032

✦ Synopsis

to cope with the voluminous and gigantic image objects.

Storage and retrieval of visual data play an important role

Since little special development is required in organizing in multimedia systems. We have developed a content-based and retrieving image objects in these systems, many image scheme for retrieving images from multimedia databases intellior pictorial databases in early years are attribute-based gently. The retrieval takes two stages. The first stage retrieves [1]. Indexing methods used in these systems are usually an image based on partial information. In the second stage, Bϩ trees [2] or inverted files [3]. Chang and Knuii [4] the system accumulates knowledge from the results of the firstsurveyed a variety of attribute-based pictorial database stage retrieval. It analyzes the subspace of features from the systems. The main problem in attribute-based systems is resulting images and tries to understand the query request. It that the adoption of a fixed set of attributes cannot take also makes full use of the entire index space, although queries advantage of the rich content in images. Moreover, these can be made on partial information. The technology developed attributes would limit the scope of the application of datawill find many applications in multimedia areas. It will also provide a tool for studying how humans rank the similarity of bases and provide little room for future and unforeseen images and what information people use in visual perception, usage of the images. etc., and will help in the development of methods based on Text-based methods annotate images with a text descripthese human approaches. © 1996 Academic Press tion and structured fields. Queries in these systems use keywords and return partially matched results ranked using similarity. Indexing methods for text-based image systems