A Zero-Watermarking Scheme using Discrete Wavelet Transform

Abstract

Digital watermarking is an efficient technique to protect copyright and ownership of digital information. But in traditional methods of digital image watermarking, the texture of original image gets distorted more or less. In this paper, a zero-watermarking based copyright protection scheme is proposed. Watermark is not embedded in the host image rather it is encrypted with the host image. Two schemes are proposed, the basic idea behind both the schemes is making use of Discrete Wavelet Transformation and Singular Value Decomposition to extract robust features of host image. The first approach consists of dividing the host image into overlapping sub images of size 8 × 8. Each block is further decomposed up to level-one using Discrete Wavelet Transform followed by Singular Value Decomposition. In the second approach the image is first subject to DWT, then the approximation part is chosen and later it is divided into overlapping sub-images of size 4 × 4. The further steps are same for both the algorithms. The encryption of the secret image with the host image is done by generating two shares: the master share and the ownership share. The master share is generated by differential classification of features extracted. The ownership share is generated with the help of master share and secret image. The two shares separately do not give any clue of the secret image but when put together they reveal the encrypted secret image. Experimental results show how the proposed algorithm is successful in resisting some common image processing and geometrical attacks.