An Algorithm of Mean Quantification Digital Watermarking based on Lifting Wavelet Coefficients

Abstract

Because traditional wavelet transformation is decimals, which inevitably introduce transformation error in image watermarking. A digital watermarking algorithm is presented based on the mean quantification of the second generation wavelet. After a host image is decomposed by the lifting wavelet, its low frequency coefficients is used as performing the mean and quantification processes, and embedding and extracting the blind watermark within it. Experimental results show that the algorithm has a fine transparence of embedding watermarking and has strongly robust to usual image process multi-attacks.