A new blind image watermarking method based on shur transformation

As the internet becomes more and more populous, people concern more about ownership and intellectual rights protection. Digital watermarking becomes today an efficient way of securing multimedia documents. In the recent years, a lot of papers discussed the ways to embed watermark into the image and many techniques are proposed in the field. These techniques depend on the representation domain of the image (spatial, frequency and multi-resolution). Every domain has its specific advantages and limitations. Each technique in a chosen domain is found to be robust to specific sets of attacks types. Also, while the image channel that will hold the watermark can be transformed to many different domains, therefore it has different capacities and probability of errors with respect to these different transformations. These already used old domains are deeply explored and the attacks to defeat the developed techniques in these domains are from day to day built-up. In this paper, a new watermarking method is presented using a new domain for the image representation and watermark embedding. This domain based on the mathematical SCHUR transformation, is found to be robust against many sets of STIRMARK attacks such as lossy JPEG compression, convolutions filtering, noise addition and other image processing attacks. Using this new domain technique, the robustness to preserve and extract the embedded watermark is improved with respect to different other methods. The experiment results show better and efficient embedding technique and lower probability of error detection. In addition the use of the host image is not needed in the watermark detection process.