An Efficient Region Matching Method for Content-based Image Retrieval

Content-based image retrieval, which provides convenient ways to retrieve images from large image databases, has been studied actively. However, many previous image retrieval techniques do not look at regions in an image. In this paper, we present a novel region matching method for contentbased image retrieval. In this method, the region dependences in each image are first computed in order to select the most interesting region from each image. Further, the distance of regions between different images is defined. Finally, the similar-ity measurement formulation is proposed. Experimental results show that the proposed method can improve the precision rate and the recall rate of image retrieval.