A semi-blind image watermarking based on Discrete Wavelet Transform and Secret Sharing

Abstract

In this paper the concepts of frequency domain watermarking and Secret Sharing are combined to protect the copyrights of a digital image. The proposed method employs the concept of XOR-based Visual Secret Sharing (XVSS) to split a watermark into two unexpanded parts, a public watermark and a private watermark. The private watermark is extracted from the features of host image and original watermark. The public watermark can be extracted from the controversial image at any time without the need for original host image and/or watermark. Since private watermark is needed in recovering the original watermark, the proposed scheme comes under the category of semi-blind watermarking. The proposed watermarking technique aims at improving the security of similar watermarking schemes by processing LL sub-band coefficients of Discrete Wavelet Transform (DWT) of the host image to satisfy central limit theorem. The simulation results reveal that the proposed method can resist to a variety of common image processing manipulations.