Integrating symbolic images into a multimedia database system using classification and abstraction approaches

Symbolic images are composed of a finite set of symbols that have a semanti c meaning. Examples of symbolic images include maps (where the semantic mea ning of the symbols is given in the legend), engineering drawings, and floo r plans. Two approaches for supporting queries on symbolic-image databases that are based on image content are studied. The classification approach pr eprocesses all symbolic images and attaches a semantic classification and a n associated certainty factor to each object that it finds in the image. Th e abstraction approach describes each object in the symbolic image by using a vector consisting of the values of some of its features (e.g., shape, ge nus, etc.). The approaches differ in the way in which responses to queries are computed. In the classification approach, images are retrieved on the b asis of whether or not they contain objects that have the same classificati on as the objects in the query. On the other hand, in the abstraction appro ach, retrieval is on the basis of similarity of feature vector values of th ese objects. Methods of integrating these two approaches into a relational multimedia da tabase management system so that symbolic images can be stored and retrieve d based on their content are described. Schema definitions and indices that support query specifications involving spatial as well as contextual const raints are presented. Spatial constraints may be based on both locational i nformation (e.g., distance) and relational information (e.g., north of). Di fferent strategies for image retrieval for a number of typical queries usin g these approaches are described. Estimated costs are derived for these str ategies. Results are reported of a comparative study of the two approaches in terms of image insertion time, storage space, retrieval accuracy, and re trieval time.