Enhancement of image retrieval by image attention based on edge features in a multiresolution grid framework

Abstract

Region-based image retrieval (RBIR) is widely used for enhancing the performance of content based image retrieval (CBIR). In this paper, we present a novel scheme to enhance retrieval performance using image attention analysis based on edge features within a multiresolution grid framework. The coarse segmentation by the grid, yielding image tiles, will serve as basic building blocks for feature computation. The edge image is computed using gradient vector flow (GVF) field. The image is partitioned into subimages based on the edge image. An integrated subimage matching scheme is used for image similarity. The experimental results demonstrate the efficacy of the method. © 2007 IEEE.