An Improved Blind Watermarking Method Based on SWT and LU Decomposition

In order to improve the security of traditional digital image watermarking algorithm, and solve the problem of digital watermark is sensitive to signal processing and geometric distortion. A new digital watermarking algorithm combined with stationary wavelet transform (SWT) and LU decomposition is proposed. Unlike the common transform domain watermarking techniques, this scheme decomposes the whole image with SWT firstly, and then divides the image into 8 x 8 blocks, taking SVD decomposition for each block and selecting an embedded position of each block to add watermarking. Here, instead of embedding the watermark value directly, we decompose the image by LU and embed the key values into S matrix. By calculating the average of all the embedded position value, and comparing each embedded position value with average value, getting the key which is used to realize the blind detection. Experiments show the blind watermarking algorithm is proposed in this paper not only has good fidelity, but also has good robustness to various attacks.