Audio Watermarking for Copyrights Protection

ABSTEACT Digital watermarking is the process that imperceptibly watermarks the multimedia product with a specific watermark for the purpose of content authentication, data monitoring and tracking, and copyright protection. Amongst the applications, the most prominent usage of watermarking technique is helping in identifying the origins of different multimedia files and resolving ownership disputes. Specifically, the report is concerned with audio watermarking for copyright protection. In the report, issues on audio watermarking for copyright protection are discussed in details. Firstly, introduction to audio watermarking is made in section 1, which is a glance of the state-of-the-arts in audio watermarking technique. Secondly, section 2 describes the principles of Psychoacoustics. Psychology of hearing, especially masking phenomenon, and psychoacoustic model are included. Thirdly, requirements of audio watermarking for copyright protection are stated in section 3, such as imperceptibility, robustness, security, capacity and speed. Actually, these prerequisites are what subjective and objective evaluations rely on. Section 4 is the extension of section 3. In the design of any audio watermarking scheme, robustness is a typical aspect worthy of considerations. So section 4 studies on common signal processing operations and deliberate manipulations, in order to get a better understanding of the mechanisms of various attacks. It is help for performing correct detection. Next, section 5 and 6 give detailed insight into different approaches on audio watermarking and also some supplemental techniques. Each basic system, i.e. LSB modification, phase coding, spread spectrum watermarking, cepstrum domain watermarking, wavelet domain watermarking and echo hiding, is implemented and evaluated, so as to compare their performances and find out their strengths and weaknesses. Based on the studies above, a robust and secured audio watermarking system using Gammatone auditory filterbank, coded image, multiple scrambling and adaptive synchronizing is developed in Section 7. Its embedding and detection methods are expounded and a complete assessment is given. Finally, Section 8 concludes the report and literature cited are listed in the last section.