An intelligent watermark detection decoder based on independent component analysis

The detection errors undermine the credibility of a digital watermarking system. It is crucial to have a reliable detector such that false detection rate can be minimized before digital watermarking can be widely used in the real world. An intelligent watermark decoder using Independent Component Analysis (ICA) is proposed in this paper. The mathematical model of the detector is given in this paper. This intelligent decoder extracts a watermark correctly without using the original image. However, the accuracy of the watermarking extraction depends on the key and the statistical independence between the original image and the watermark. Experimental results have demonstrated that the proposed intelligent watermarking technique is robust against the attacks produced by Stirmark, such as cropping, filtering, image compression, rotation, scaling. It also has a good robustness against combination of several kinds of attacks. It is indicated that this new intelligent decoder has superior advantages over the existing ones in many aspects. © Springer-Verlag 2004.