摘要
提出了一种DWT-SVD域的自适应ICA水印算法。首先对原始图像进行小波变换,分解出其逼近子图,对以低频信息为主的逼近子图再进行奇异值分解,然后根据逼近子图的噪声可视函数,用独立分量分析方法在不同区域嵌入不同强度的水印信息,达到自适应的目的。实验结果表明,水印信息在噪声干扰、图像处理、图像压缩下具有很好的鲁棒性,同时能满足水印的不可见性。
The paper proposes an adaptive watermarking algorithm based on DWT-SVD domain.The original image is transformed into discrete wavelet domain.The watermark is embedded into the singular value decomposition of the approaching sub-image with different strength according the noise visibility function of the sub-image by the independent component analysis.Experimental results show that the algorithm has good robustness against noise, normal image processing, image compression.
出处
《微计算机信息》
2010年第8期195-196,共2页
Control & Automation
关键词
离散小波变换
奇异值分解
噪声可视函数
独立分量分析
discrete wavelet transform
singular value decomposition
noise visibility function
independent component analysis
