Semantic clustering and querying on heterogeneous features for visual data

The effectiveness of the content-based image retrieval can be enhanced using the heterogeneous features embedded in the images. However, since the features in texture, color, and shape are generated using different computation methods and thus may require different similarity measurements, the integration of the retrieval on heterogeneous features is a non-trivial task. In this paper, we present a semantics based clustering approach, termed SemQuery, to support visual queries on heterogeneous features of images. Using the approach, the database images are classified based on their heterogeneous features. Each semantic image cluster contains a set of subcluster that are represented by the hetroogeneous features that the images contain. A database image is included into a feature subcluster only if the image contains all the features under the same cluster. We also designed a multi-layer model to merge the results of basic queries on individual features. A visual query processing strategy is then presented to support visual queries on heterogeneous features. Experimental analysis is conducted and presented to demonstrate the effectiveness and efficiency oft he proposed approach.