Research And Implement Of Digital Audio Watermarking

With the rapid development of the network (especially Internet) and multimedia technique, the protection of intellectual property rights has been the key problem which we must solve. In recent years, MP3 music has become extremely popularized owing to its advanced coding techniques, however, with the development of digital networks and multimedia processing technology, MP3 music is much easier to be edited, duplicated and distributed, leading magnitude economic losses to the owner of the music. Therefore, the copyright protection,information security and authentication of the digital audio has been a focus in network digital world. Traditional encryption techniques are insufficient in security and in communication. Digital watermarking technology, considered as an effective complement solution for traditional encryption method, is a new technique for copyright protection and integrality authentication in the open network environment. Digital watermarking technology is becoming an international academic focus.This dissertation focuses on research on watermarking algorithms of digital audio. The main content includes:(1)A semi-fragile audio watermarking algorithm for copyright protection and content authentication is proposed. The features of the proposed algorithm are as follows:⑴T he original audio signal can be adaptively cut, and the watermark depth can be intelligently adjusted according to human auditory masking and local audio character.⑵T he multi-resolution characteristics of DWT and the energy-compression characteristics of DCT are combined to improve the transparency of digital watermark.⑶The copyright protection and content authentication for digital audio can be utilized simultaneously.(2)On the basis of support vector regression (SVR), a new adaptive digital audio blind watermarking algorithm is proposed. This algorithm embeds template information and watermark signal into the original audio by adaptive quantization according to the local audio correlation and human auditory masking. In the procedure of watermark detection, Firstly, extract the corresponding feature of template and watermark from watermarked audio; then select the corresponding feature of template as strain sample, the SVR strain function can return a model; Finally, according to the corresponding feature of watermark, the model can be used for future prediction to exactly recover the watermark from the watermarked audio.(3)Synchronization attack is one of the key issues of digital audio watermarking. In this paper, a robust digital audio watermarking algorithm is presented, which can resist synchronization attack effectively. The features of the algorithm are as follows:⑴More steady features extraction and new embedded strategy are adopted to resist the synchronization attack effectively.⑵The multi-resolution characteristics of DWT and the energy-compression characteristics of DCT are combined to improve the transparency of digital watermark.⑶The algorithm can extract the watermark without the help of the original audio signal.