摘要
为进一步提高水印抵抗攻击的能力,提出一种基于量化的混合提升小波变换和奇异值分解的自适应数字图像水印算法。对二值水印信号进行异或加密与置乱处理,以增强水印信号的安全性和鲁棒性;对原始载体图像进行互不重叠的分块,对随机选取的分块进行一级提升小波变换,对生成的低频子带进行奇异值分解;根据水印信号的取值,对每一分块低频子带的最大奇异值采用自适应量化的方法嵌入水印信号。仿真结果表明,该算法具有较好的透明性和抵抗常规信号处理攻击的能力。
To further enhance the ability of the watermarking to resist attacks ,a adaptive digital image watermarking algorithm based on quantization in lifting wavelet transform （LWT） and singular value decomposition （SVD） domain was proposed .First‐ly ,to enhance the safety and robustness of watermarking ,the binary watermarking signal was preprocessed using XOR encryp‐tion and Arnold transform .Secondly ,original image was subdivided to non‐overlapping blocks and 1‐level lifting wavelet trans‐form was performed for each selected blocks ,and singular value decomposition was applied to the low frequency sub‐band .Final‐ly ,according to the value of the binary watermarking signal ,the biggest singular value of low frequency sub‐band was modified by exploiting adaptive quantification method to embed watermarking .Experimental results show that the proposed algorithm has good transparency and it is robust against common signal processing .
出处
《计算机工程与设计》
北大核心
2015年第6期1494-1497,1503,共5页
Computer Engineering and Design
基金
国家自然科学基金项目(61309006)
贵州省科学技术厅
贵州师范大学联合科技基金项目(黔科合LH字[2014]7041号)
2013广东省千百十高层次人才培养基金项目(粤财教[2013]246号)
关键词
图像水印
奇异值分解
提升小波变换
量化
鲁棒性
盲提取
image watermarking
singular value decomposition
lifting wavelet transform
quantization
robustness
blind ex-traction
