Digital watermarking for audio content protection

• Main Author: Lei, Baiying
• Other Authors: Soon Ing Yann
• Format: Theses and Dissertations
• Language: English
• Published: 2012
• Subjects: DRNTU::Engineering::Electrical and electronic engineering
• Online Access: https://hdl.handle.net/10356/50758
• Institution: Nanyang Technological University

Rapid development of the digital technology and internet has made copyright protection and ownership verification of digital audio data hot research topics that have attracted a lot of interest from different parties. Since the traditional audio protection algorithms cannot solve all the problems of audio copyright protection and digital right management (DRM) due to the unique audio data features and algorithm limitations, this thesis is proposed to address the issue of audio copyright protection using watermarking techniques. Firstly, a new perception-based watermarking scheme for MPEG-4 scalable to lossless (SLS) audio is proposed. Under the control of the psychoacoustic model, the significant part of integer modified discrete cosine transform (IntMDCT) coefficients are adaptively modified during MPEG-4 SLS audio compression taking robustness, imperceptibility and security into consideration. The chaotic watermark generated by chaos is then embedded to ensure security. Moreover, the adaptive spread spectrum (SS) method is exploited to further optimize the tradeoff between robustness and transparency of this scheme. Secondly, a blind and robust audio watermarking scheme based on singular value decomposition (SVD) – followed by discrete cosine transform (DCT) with the synchronization code technique is proposed. A binary watermark is embedded into the high frequency band of the SVD-DCT block. Chaotic sequence is adopted as the synchronization code and inserted into the host signal. Thirdly, a new and robust audio watermarking scheme based on lifting wavelet transform (LWT) and SVD is proposed. Specifically, the watermark data is efficiently inserted in the low coefficient of the frequency subband of LWT audio taking advantage of both SVD and quantization index modulation (QIM). Furthermore, the synchronization code technique is also integrated in the hybrid LWT-SVD audio watermarking method. Finally, a particle swarm optimization (PSO) and SVD based audio watermarking scheme is proposed. The watermark is embedded by modifying singular values (SVs) of the host audio signal with the SS method. The designed optimization technique employs PSO to search for the optimal watermark strength by continuously adjusting the fitness values so that the conflicting requirements of robustness and imperceptibility of the watermark are satisfied optimally. Moreover, the proposed scheme has the flexibility to include or remove various watermarking attacks during training for optimization. The experimental results validate the practicality of the proposed methods and show that the proposed algorithms are very robust to typical signal manipulations. Furthermore, the results of comparison with other similar state-of-the-art audio watermarking techniques indicate that the proposed algorithms are more robust against most common signal processing attacks while preserving good imperceptibility.