Extended concept-based image retrieval system (E-COIRS)

Abstract

In this paper, we design and implement E-COIRS enabling users to query with concepts and image features used for further refining the concepts. For example, E-COIRS supports the query 'retrieve images that a black home appliance is to north of reception set'. The query includes two types of concepts: IS-A and aggregation-'home appliance' is an IS-A concept, and 'reception set' is an aggregation concept. For evaluating such a query, E-COIRS includes three Important components: a visual image indexer, thesauri and a query processor. Each pair of objects in an image captured by the visual image indexer is converted into a triple. The triple consists of two object identifiers (oids) and their spatial relationship. All the feature of an object is referenced by its old. The thesauri, which are mainly used by the query processor to detect concepts, consist of a triple rule-based thesaurus and a term thesaurus. The query processor obtains an image set associated with each triple in a user query by looking up an inverted file and CS-Tree. To support efficient storage use and fast retrieval on high-dimensional feature vectors, E-COIRS uses a new Cell-based Signature tree. E-COIRS is a more advanced content-based image retrieval system than other systems which support only concepts or image features.