Research On The Adaptive Audio Watermarking Technology For Copyright Protection In The Transform Domain

With the rapid development of the computer network technology, the transmission and storage of multimedia technology have made a lot of progress, it is also an important way for people to get information. But at the same time, malicious tamper, illegal copy and transmission of the digital multimedia information are getting easier, which has also damaged author’s copyright and economic interests seriously. So, how to protect the copyright and content the authentication of digital multimedia information effectively has become an international research hot spot. Therefore digital watermarking technology which is an effective solution in the flied of the copyright protection has obtained great attention. As an important branch of digital watermarking, the audio watermarking technology is expected to further research.The audio watermarking algorithm that used for the copyright protection should satisfy with some of the requirements such as imperceptibility and robustness. The position or strength of embedding is adaptive, which realizes the improvement of imperceptibility in the watermarking algorithm. Audio watermarking algorithms are improved from the time domain to transform domain and hybrid domain, which also have strengthened the robustness of the algorithms. This paper mainly focuses on designing transform-based adaptive audio watermarking algorithm for copyright protection, the research work is given as follows:1、An robust audio watermarking algorithm has been designed that is used for copyright protection in the complex cepstrum domain. According to the characteristics that cepstrum coefficients after complex cepstrum transform have less influence on the conventional signal processing attacks. The algorithm makes use of the method of statistical mean modulation to embed the watermark into cepstrum coefficients. The selection of watermark embedding position combines with the human auditory properties and the energy characteristics in the time domain, which makes the algorithm have better imperceptibility. The experimental results show that this proposed algorithm is robust against some conventional signal processing attacks such as adding noise, low-pass filtering, re-sampling, re-quantization, MPEG-1Layer III (MP3) compression. Therefore, the proposed algorithm has strong robustness.2、A new audio watermarking algorithm has been designed based on Lifting Wavelet Transform (LWT) and Discrete Cosine Transform (DCT), and the watermark is embedded in the hybrid domain. The algorithm mainly combines the wavelet multiresolution analysis property with the energy-compression and the relativity properties of DCT, which realizes the process of embedding by quantifying the coefficients of low frequency. By the way, the adjustment of quantification steps is according to the audio content and ideal SNR value, both the adaptation and imperceptibility are ensured. Furthermore BCH error correcting coding technique is applied to the watermark preprocessing in order to reduce the bit error rate(BER) of watermark extraction; The extraction of watermark is blind without using the original audio or original watermark. In addition, this proposed algorithm has strong resistance against some conventional signal processing attacks such as adding noise, low-pass filtering, re-sampling, re-quantization, echo, MPEG-1Layer III (MP3) compression and some attacks by StirMark software. Therefore the algorithm’s robustness has also been ensured.