Exploiting Semantic Connectivity in Redefined Data Representation for Image Retrieval

摘要

The state-of-the-art in automatic image annotation research includes many methods of combining the visual features and text to deal with the semantic gap between low-level visual feature and high-level concept. However, the text combined with images have not been clearly defined, which people disable to retrieve the desired resources by describing the queried target. Consequently, we propose a RDF-based annotation linked with ontology as DBpedia to have more semantic meanings. Although there are few image annotation researches based on ontology, no public datasets are released. In this paper, we develop an ontology-based image annotation and retrieval tool, namely OSIA. In the proposed scheme, the image dataset containing RDF annotation is announced.