摘要
根据人类视觉特征提出了一种基于小波系数相关性的自适应水印嵌入算法.该算法先对原始图像进行的小波分解,选取水印所要嵌入的子带并将所选取的子带划分成2×2大小的块,同时利用每一个小波块中相邻的4个系数之间具有一定的相关性与所要嵌入的水印图像来构造该小波块的水印系数,利用人类视觉模型调整水印的嵌入强度,从而在保证水印不可见性的基础上最大限度的保证水印的鲁棒性.选取不同特点的两幅图片进行实验,实验结果表明,该算法对高斯噪声、椒盐噪声、JPEG压缩、剪切、中值滤波等攻击都具有较好的鲁棒性,对旋转几何攻击具有一定的鲁棒性.
An adaptive watermark algorithm based on the relativity of wavelet coefficients proposed. Firstly, the original image and the watermark image are decomposed by wavelet transform; the second step is that selecting the sub-bands of embedding watermark and dividing the sub-bands into blocks, the size of block is 2×2; the next step is to construct the watermark of the block according to the relativity of four consecutive wavelet coefficients in each block. Finally, the watermark embedding strength is regulated by the human visual system model, therefore, the method gained robustness and imperceptibility. Two different images are selected to embed watermarking in the paper, according to the experimental results, the algorithm can withstand attacks such as JPEG compression, adding Gaussian noise and Salt and Pepper noise, cropping and so on. At the same time, the algorithm can withstand rotation attack at some degree.
出处
《武汉大学学报(理学版)》
CAS
CSCD
北大核心
2007年第3期309-313,共5页
Journal of Wuhan University:Natural Science Edition
基金
国家自然科学基金(10477007/A07)
湖北省教育厅重点基金(2003A012)
中国科学院信息安全国家重点实验室开放课题基金(03-02)资助项目
关键词
数字水印
小波变换
相关性
鲁棒性
人类视觉系统
digital watermark
wavelet transform
relativity
robustness
human visual system (HVS)