A Double-blind Audio Watermarking Algorithm

With the development of digital technology and internet, digital watermark technology becomes one kind of effective protection technology for digital product copy right and maintenance technology for data security, and it is one of the branches of research aspect of information hiding technology.First we review the development history of digital watermark, introduce some basic issues and conclude the framework of embedding and detection system. Secret information can be embed into audio files with these techniques, so as to achieve copyright protection and cover communication about digital watermarking and Wavelet analysis and then focused on the principles of those techniques.According to the theory presented above, a novel dual watermarking of digital audio scheme is proposed in which robust and fragile watermarking are simultaneously embedded for copyright protection and content authentication. First the robust watermark is embedded by quantizing amplitude of DWT low frequency coefficients of digital audio. Then, singular value decomposition is used to extract the DWT high frequency coefficients value that embedded fragile watermark based on modular arithmetic. And embeds watermarks into original signals with out modifying the audio data. Experimental results show that the watermarking scheme is indeed superb in terms of robustness and fragility.At last, a conclusion of this paper is given, and a prospect for the future of audio digital watermarking is also presentedThe new audio digital watermarks algorithm presented in this thesis can get better results and better trade off between the robustness and imperceptible. It is a very significant attempt for researching on the protection of audio copy right.