A hybrid SVD-DCT watermarking method based on LPSNR

In this letter, a new watermarking method which combines the singular value decomposition (SVD) and the discrete cosine transform (DCT) is presented. We consider that more transparency can easily be obtained in that only the singular values (SVs) of the watermark need to be embedded into the original image. Moreover, since the bigger SVs which comprise most energy of the watermark are embedded into the DC components of the sub-blocks of the original image, our method would be quite robust. In addition, the local peak signal-to-noise ratio (LPSNR) is adopted in our method to achieve the highest possible robustness without losing the transparency. (C) 2004 Elsevier B.V. All rights reserved.