Object-based image retrieval system using rough set approach

Abstract

In this chapter, we present an object-based image retrieval system using the rough set theory. The system incorporates two major modules: Preprocessing and Object-based image retrieval. In preprocessing, an imagebased object segmentation algorithm in the context of the rough set theory is used to segment the images into meaningful semantic regions. A new object similarity measure is proposed for the image retrieval. Performance is evaluated on an image database and the effectiveness of proposed image retrieval system is demonstrated. Experimental results show that the proposed system performs well in terms of speed and accuracy. © Springer-Verlag Berlin Heidelberg 2012.