A Robust Digital Audio Watermarking Algorithm

Aiming at the problems of insufficient embedding strength and single attack mode, this paper proposes a robust digital audio watermarking algorithm for binary images. The watermark encryption, embedding, attack, extraction and decryption are completed. The watermark image is encrypted by Arnold transform, the feature sequence is extracted by wavelet transform, and Discrete Cosine Transform (DCT) and Singular Value Decomposition (SVD) are performed. Finally, the watermark image is embedded into the audio signal. Watermark extraction is the reverse operation of watermark embedding. Arnold inverse transform is applied to the extracted watermark to obtain the decrypted watermark image. Low-pass filtering, random noise, cropped audio, and resampling are used for the attacks. Normalized Correlation coefficient (NC), Symbol Error Rate (SER), Pearson Correlation Coefficients (PCCs), and Signal-to-Noise Ratio (SNR) are used for objective evaluation. Experiment results show the proposed algorithm has better robustness and invisibility, which is favorable for the privacy protection of audio information.