Abstract

Considering the fact that most of the multimedia database systems include querying by object features, a novel approach for color and shape querying and retrieval is proposed. The approach is histogram-based and it is supported by an auxiliary object extraction tool in the query-by-feature component of a multimedia database system. The object extraction tool is semi-automatic, hence it can successfully capture salient object regions for most of the images and/or video frames. The use of the object extraction tool facilitates object-based querying of color and shape content. In the histogram-based approach, the color and shape can be integrated to improve the performance of querying. It is shown through performance experiments that the histogram-based approach overcomes the deficiencies of the existing methods for querying by shape. The evaluation of the effectiveness and the robustness of the histogram-based approach is also presented via performance experiments.

Citations