A robust digital audio watermarking scheme based on DWT and Schur decomposition

TLDR: The proposed digital audio watermarking scheme based on a discrete wavelet transform and Schur decomposition hybrid method is inaudible and robust against common types of attacks such as Gaussian noise, re-quantization, re -sampling, low-pass filter, high- pass filter, echo, MP3 compression, and cropping.

Abstract: The vulnerability of digital audio signals for different types of risks requires an imperceptible and robust digital audio watermarking scheme. In this research, we propose creating one such imperceptible and robust hybrid watermarking scheme based on a discrete wavelet transform (DWT) and Schur decomposition hybrid method. The proposed scheme embeds the foreground bits of the watermarking image into the least significant bit of the diagonal coefficients of the triangular matrix S generated from Schur decomposition. Schur decomposition is applied on the second sub-band HL2 generated from applying a second-level 2D-Haar DWT on the first channel of the original audio signal. We analyze the proposed digital audio watermarking scheme’s performance in terms of signal to noise ratio (SNR), objective difference grades (ODG), and subjective difference grades (SDG) that resulting 81.43, 4.78 and 0.184, respectively. The resulting of payload capacity, NC, and BER are as high as 319.29 bps, 0.9911, and 0.0135, respectively. Experimental results confirm that the proposed scheme is inaudible and robust against common types of attacks such as Gaussian noise, re-quantization, re-sampling, low-pass filter, high-pass filter, echo, MP3 compression, and cropping. In comparison with state-of-the-art audio watermarking schemes, the proposed scheme’s performance is superior in term of imperceptibility, robustness, and data payload size.