Abstract: In this paper, an adaptive watermarking algorithm of image based on Singular Value Decomposition (SVD) is proposed. Binary barcode is taken as watermarking embedded into the S component of SVD. By exploiting the visual masking and texture characteristics of Human Visual System (HVS), the strength of embedded watermarks is decided by the gray and texture of the original image. It ensures the self-adaptation of the intention of the embedded watermarks to the original image. The watermarked image is adjusted by using Radon transform before the watermarking is extracted and then the extracting barcode is corrected according to the characters of itself. The experimental results show that the proposed algorithm is invisible and robust against many commonly used image-processing methods and especially against geometric attack. This technology will be of good application in printing security.

Key words: digital watermarking, singular value decomposition (SVD), barcode, Radon transform