IMAGE RETRIEVAL BY COLOR SEMANTICS WITH INCOMPLETE KNOWLEDGE

Retrieval by content from image databases faces the distance between l ow-level syntactic features that can be automatically detected by conv entional image processing tools and high level semantics which capture s user's filtering intentions. A system is presented which bridges thi s gap by resorting to a theory formulated by Johannes Itten in 1960, a nd widely accepted in the community of fine arts, to support objective interpretation of color arrangements over paintings. The system relie s upon a schema distinguishing archiving, querying, and retrieval stag es. In the archiving stage, images are associated with a description c apturing the spatial arrangement of regions with homogeneous chromatic attributes, as detected by the use of an automatic image processing t ool. imprecise descriptions are supported through the adoption of a hi erarchical index providing a multi-resolution representation of image contents. In the querying stage, a visual iconic language allows the e xpression of sentences about chromatic contents in accordance with a h igh-level semantic model of colors combinations. By permitting flexibl e expression of abstract, non-literal, properties, the model supports intentional vagueness and incompleteness in the specification of searc hing queries. In the retrieval stage, a similarity score is introduced , which accounts for the degree with which a query assertion applies t o a given image. The measure of similarity drives the traversal of the hierarchical index up to find the minimum level of description precis ion, permitting a definite decision about the satisfaction of the quer y on each stored image.