Skip to Main Content
Content-based image retrieval (CBIR) is one of the most exciting and fastest-growing areas in the field of multimedia technology. In this Letter, we apply the watermarking technique into the retrieval system and propose a novel approach for JEPG image retrieval. The proposed image retrieval system consists of two main phases, offline process and online retrieval process. The feature vector is extracted from each image as the watermark to be embedded into the image, which is the preprocessing operation called offline process. It doesn't require decompressing the JPEG images but directly embedding the watermark in the DCT domain. The online retrieval process consists of three processes, i.e., query feature computation, watermark extraction and feature vector matching. Since the features are embedded in the image data, it is unnecessary to compute the features but only to extract it from the watermarked image. We carry out a series of experiments on a watermarked image database, and the detailed comparison results between the novel approach and the traditional system indicate the advantage of the proposed method.