Optimized model embedding invisible robust watermark

To set the proper strength of the embedded watermark signal is a critical problem for obtaining a robust and transparent watermark in image watermarking. In this paper, to solve this problem, a new simple scheme on how to optimally embed a watermark with fidelity and robustness constrains is proposed. Firstly, human visual system (HVS) model is used to get masking function by calculating the just noticeable distortion (JND) mask of image directly in the spatial domain. Then, based on adaptive iterative algorithm, the strength of the embedded watermark in discrete cosine transform (DCT) domain attains to maximum little by little under the rule of masking function. The new scheme has been tested with attack software StirMark 4.0. Compared with the conventional scheme, simulation results show that more robust watermark can be attained while retaining high image quality.