基于量化和听觉感知的数字音频水印

Audio Watermarking Based on Quantization Index Modulation and Auditory Perception

论文提出了一种新的用于音频作品版权保护的数字水印算法。水印(二值标志图像)由相邻小波包系数的特定量化关系表示。心理声学模型控制水印嵌入和小波包分解过程,确保水印嵌入是不可感知的,同时使子带分解结构接近于临界频带,因此与以往算法相比该算法提高了子带掩蔽阈值,即加大了水印的抗攻击范围,增强了水印的鲁棒性。实验结果显示该算法能有效抵抗MP3压缩,低通滤波,重新采样和添加白/随机噪声等多种攻击。

A novel audio watermarking scheme for copyright protection is proposed in this paper.The watermark,which is a binary logo image,is embedded during quantizing the wavelet packet coefficients.The psyehoaeoustie model employed in the method is the combination of the MPEG Ⅰ Psychoacoustie Model Ⅱ and wavelet packets.The best basis of the wavelet packet decomposition adapts to perceptual criterion so that the subbands are closer to Critical Bands. This makes the subband masking thresholds in the new approach larger than the ones calculated based on FFT,whieh enlarges the range of the watermark to resist attacks.Furthermore,the presentation of the watermark bits （the quantization relationships between adjacent coefficients） decreases the successful attacks chances.The presented approach is an oblivious one.Experimental results show that the proposed scheme can resist a variety of signal manipulations and attacks,including MPEG compression,low-pass filtering,resampling,white/random noise adding effectively.