Research On Multi-watermarking Based On The Model Of Communications

Multi-watermarking technology is proposed in the research of digital watermarking technology used to solve the problems of multiple ownership or copyright.. It can not only improve the robustness, but also works to improve the watermark embedding capacity. which is a more secure, efficient, and practical watermarking.Basic theory of multi- watermarking is introduced briefly in this thesis and its advantage and future trends compared with one watermark embedding is described ,which is a more practical embedding technology.Through the study of large amounts of literature data, a novel dual watermarking algorithm based on wavelet transform for color images is proposed. Firstly, transforming the RGB color image into YIQ color space, and then decomposing the Y with discrete wavelet transform. Arnold transform and error correcting coding are adopted in the watermarking processing, and then embedding them into the detail sub-image. Above all, this algorithm can also realize blind extraction. The watermark generated with proposed algorithm is imperceptible and of better robustness against common attacking processing methods. Innovative paper is mainly reflected in three aspects:1 Different pretreatment is adopted for different watermarks, front-end processing from the perspective of improving watermark robustness. Watermark (the first one) was scrambling for a deal, disrupting the correlation between the original image, so that the watermark has repeatedly against cutting attacks demonstrated strong robustness; on the second application of spread spectrum watermarking technology and error correction coding techniques,watermark in the anti-noise areas showed considerable robustness, so as to achieve enhanced overall robustness of the watermark .2 Different algorithms which are different from common embedded rules of adding ,multiplying so on are introduced to finish multi-watermark embedding. The former is based on the relationship between the coefficients of the embedded watermark; the latter is based on the mean coefficient to embed the watermark for improving the security of watermarking algorithm.3 On the embedded watermark in accordance with their respective characteristics, realized the blind extraction and detection. The advantage of double watermarking algorithm is that :as long as any of a watermark can be detected or extracted, the algorithm is effective. That is the case without the original image, you can extract or detect the watermark. Therefore it can be widely used in future.