摘要
针对现有水印算法在嵌入过程中通常会对原始音频信号产生破坏的问题,设计了一种基于离散小波变换-奇异值分解(DWT-SVD)的水印嵌入方法实现水印信号的嵌入与提取.同时,以相关系数和误码率为衡量指标,对水印算法的不可感知性、安全性及鲁棒性进行评定测试.结果表明:在经过噪声、滤波、剪切、压缩等多种攻击后,所提取的水印仍然能保持较高的清晰度,这说明DWT-SVD算法具有较强的抗攻击能力,可有效保护版权人的利益.
The existing watermarking algorithm in the embedding process usually issues the destruction of the original audio signal,and design a watermark embedding method based on the discrete wavelet transform-singular value decomposition（DWT-SVD）to realize the watermark embedding and extraction.At the same time,with the correlation coefficient and bit error rate as the measuring index,the non-sense,security and robustness of the watermarking algorithm are evaluated and tested.Results show that after noise,filtering,cropping,compression multiple attacking,the extracted watermark can still maintain a high clarity,which indicates that DWT-SVD algorithm has strong anti-attack ability,can effectively protect the copyright owner′s interests.
出处
《华侨大学学报（自然科学版）》
CAS
北大核心
2016年第6期770-773,共4页
Journal of Huaqiao University(Natural Science)
基金
国家自然科学基金青年基金资助项目(61300165)
关键词
小波变换
水印算法
音频信号
奇异值分解
wavelet transform
watermarking algorithm
audio signal
singular value decomposition
作者简介
通信作者：冯小明（1960-）,女,高级工程师,主要从事计算机技术的研究.E-mail：671003753@qq.com.
