摘要
提出了一种基于CT-CZT-SVD的数字水印算法。该算法利用CZT变换算法的高频窄带分析能力和奇异值的稳定性对CT变换获取的表示细节特征的图像高频分量进行混合处理。实验表明,该算法通过提高频率分辨率及频谱锐化效果,能充分表达图像的细节信息,有效提升了水印图像的不可见性以及对常见的几何攻击、JPEG压缩等水印攻击的鲁棒性。
We propose a digital watermarking algorithm based on CT-CZT-SVD. The algorithm uses the high-frequency-narrow-band analysis capability of CZT and the stability of singular value to deal with the high-frequency components of the image obtained by the CT. The experimental results show that the proposed algorithm can fully express the detail information of the image by increasing frequency resolution and spectral sharpening effect and effectively enhance the imperceptibility and robustness to common geometric attacks, JPEG compression and so on.
出处
《电子设计工程》
2016年第7期171-174,共4页
Electronic Design Engineering
基金
国家自然科学基金青年科学基金(61101196)
教育部高校博士点专项基金(20103219120016)
江苏省企业博士集聚计划(20110114)
关键词
轮廓波变换
线性调频z-变换
奇异值分解
鲁棒性
contourlet transform
chirp z-transform
singular value decomposition
robustness
