基于SWT和DCT的音频水印算法  

Audio watermarking algorithm based on SWT and DCT

作　　者：齐燕博 杜庆治[1] 龙华[1] 邵玉斌[1] QI Yan-bo;DU Qing-zhi;LONG Hua;SHAO Yu-bin(Faculty of Information Engineering and Automation,Kunming University of Science and Technology,Kunming 650500,China)

机构地区：[1]昆明理工大学信息工程与自动化学院,云南昆明650500

出　　处：《陕西理工大学学报（自然科学版）》2023年第1期51-59,共9页Journal of Shaanxi University of Technology:Natural Science Edition

摘　　要：为了解决在音频中添加水印信息后如何保持音频质量以及嵌入的水印信息在遭受攻击后的安全问题,提出了一种基于平稳小波变换(SWT)和离散余弦变换(DCT)的音频水印算法。首先,利用Lorenz混沌系统生成密钥对原始水印信息进行加密。然后对原始音频分帧,通过音频帧的能量特征和过零率特征确定水印嵌入帧,对水印嵌入帧进行三级平稳小波变换后,将其三级近似分量平均分成两个一维矩阵,分别进行离散余弦变换,同时计算这两个一维矩阵幅度绝对值的平均值,通过不同的水印信息修改离散余弦变换系数嵌入水印。通过实验选取3种不同类型音乐(classical、hip-hop、rock)测试该算法,结果表明,3种音乐的音频的信噪比分别为25.451 7、22.296 3、25.243 1,高于国际标准;在经过各种攻击后其误码率均在0.02以下;相关系数都在0.98以上。In order to maintain the audio quality after adding watermark information to audio and ensure the security of the embedded watermark information after an attack, an audio watermarking algorithm based on the stationary wavelet transform(SWT) and discrete cosine transform(DCT) is proposed. Firstly, the Lorenz chaos system is used to generate a key to encrypt the original watermarked information. Then the original audio is framed, and the watermark embedding frame is determined by the energy characteristics and the over-zero rate characteristics of the audio frame. After the three-level stationary wavelet transform is applied to the watermark embedding frame, the three-level approximation components are equally divided into two one-dimensional matrices, and the discrete cosine transform is applied to each of them. The signal-to-noise ratios of the algorithm for the three tested audio types(classical、hip-hop、rock) were 25.451 7, 22.296 3, 25.243 1, respectively, which is higher than the inevnational standaard;the BER was below 0.02 and the correlation coefficient was above 0.98 after various attacks.

关 键 词：平稳小波变换 LORENZ混沌系统 离散余弦变换 水印安全 

分 类 号：TP309[自动化与计算机技术—计算机系统结构]