Content-based retrieval from compressed-image databases

Summary form only given. There is an enormous amount of multimedia data including images, video, speech, audio and text, distributed among the various computer nodes on the Internet. The extent to which a user will be able to derive useful information from these data depends largely on the ease with which required data can be retrieved from the databases. The volume of the data also poses a storage constraint on the databases; hence these data will need to exist in the compressed form on the databases. We concentrate on image data and propose a new paradigm in which a compressed-image database can be searched for its contents. In this paradigm, the need for a separate index is obviated by utilising image compression schemes that can support some form of object search in the compressed be domain. The central idea is to store the image in layers of different resolutions and to be able to synthesise an edge image from a subset of the layers. This edge image then constitutes a model of the image that can be used as a searchable index. The implication of this approach is that the index is inherent in the compressed image file and does not occupy any additional storage space as would be the case in a conventional index. The preliminary results obtained from the system simulated in our experiments indicate the feasibility of the proposed paradigm.