Content-based image retrieval using salient boundary and centroid-radii model

Abstract

In view of the instability and low efficiency of the present image retrieval method, especially for simple image comparison with some salient shapes, a new image retrieval algorithm based on salient closed boundary is presented. Firstly, the Canny operator is performed to detect edges. Secondly, the ratio contour is used to extract the most salient closed boundary of some shape from the image. Finally, the similarities are measured by feature vector of the salient closed boundary based on the centroid-radii model. Preliminary experimental results demonstrate that the proposed method is quite suitable for many professional image retrieval systems and has a good performance in both retrieval efficiency and effectiveness. © 2009 Springer Science+Business Media, LLC.