Digital watermarking based on multi-band wavelet and principal component analysis

This paper presents a robust watermarking scheme based on multi-band wavelet and principle component analysis (PCA) technique. Incorporating the PCA technique, the developed blind watermarking in multi-band wavelet domain can successfully resist common signal processing such as JPEG compression with quality factor as low as 15, and geometric distortion such as cropping (cropped by 50%). Different from many other watermarking schemes, in which the watermark detection threshold is chosen empirically, the false positive of the proposed watermarking scheme could be calculated, so watermark detection threshold could be chosen based only on the target false positive. Comparing with similar watermarks in conventional two-band wavelet domain, greater perceptual transparency and more robustness could be achieved for the proposed watermarking scheme. The parameterized multi-band wavelet also leads to a more secure embedding domain, which makes attacks more difficult.