An audio watermarking scheme using discrete fourier transformation and singular value decomposition

This paper introduces an audio watermarking scheme using discrete Fourier transformation (DFT) and singular value decomposition (SVD) for copyright protection of sound contents. In our proposed scheme, initially the original audio is segmented into non-overlapping frames. Spectrum of each frame is computed by DFT. Prominent spectral peaks are detected from the frequency spectrum of each frame using a peak detection algorithm. SVD is applied to the selected prominent peaks of each frame represented in a matrix form. Watermark is then embedded into the highest singular value of each diagonal matrix. Watermark is extracted by performing the inverse operation of watermark embedding process. Simulation results indicate that the proposed watermarking scheme outperforms conventional methods in terms of imperceptibility and efficiency, while maintaining comparable robustness against various kinds of attacks such as noise addition, cropping, re-sampling, re-quantization, low-pass filtering, and MP3 compression.