Digital Audio Zero-Watermarking Algorithm And Its Research

As the rapid development of the socity, digital multimedia products have been widely used. But digital products can easily be copied and modified, piracy and manipulate phenomenon becomes spread. Therefore how to protect copyright and legal rights become a hotspot in the research through the digital watermarking technique effectivelyMain contents of this thesis are as follows:(1)Meaningless zero-watermarking in the scheme was based on wavelet decomposition and the stability of the introduction of singular value, and the test results showed that the algorithm compared to other general algorithm has better robustness. When meaningful zero-watermarking was constructed, the scheme adopted dual chaotic algorithm to encrypt system. The algorithm used RBF to construct meaningful zero-watermarking, training audio watermarking signal as the target. Under the same experimental environment compared to BP network, the results show that RBF network have better robustness. Whether meaningful or meaningless zero-watermarking algorithm in the thesis, do not change the perception of the original audio quality and has good resistance to conventional attack capability. The scheme also realizes a blind detection.(2)As the past dual watermarking has poor robustness, low signal to noise rate and inaccurate positioning, a novel dual watermarking based on wavelet packet was presented. First the robust watermark possesses strong robustness to resist attacks, which was constructed by taking advantage of the stability of singular value decomposition(SVD) and the security of chaotic system. Then a new bipolar quantization was used to enhance the sensibility of fragile watermark. The scheme effectively improved signal to noise rate attributing to the odd coefficients by quantizing the difference of the sum of the odd and the even coefficient. The experimental results show that the robust watermark provides high robustness to conventional attacks, and that the fragile watermark has very sensitive to malicious tampering which can be able to distinguish conventional attacks and malicious tampering. It also has good accuracy of tamper localization.