摘要
提出了一种基于近邻插值取样法的自适应盲检测彩色图像水印算法.该算法利用近邻插值取样法,根据载体图像的大小,自适应调节二值水印图像的尺寸,经Arnold置乱预处理后,嵌入到彩色载体图像蓝色分量的DCT高频系数中,检测时不需要原载体图像.实验表明,该算法具有较强的鲁棒性和良好的不可感知性,可有效低抗剪切、加噪、锐化等常见的图像攻击.
In this paper,an adaptive watermarking algorithm based on Nearest Neighbor algorithm of color image is proposed. Modifying the binary image according to the size of the cover image by utilizing Nerest Neighbor and through Arnold shuffle pretreatment, the algorithm embed the binary image into the high frequency band of the blue component of the color image' s DCT coefficients. The watermarking can be extracted without the original image. Experimental results show that the proposed watermarking algorithm is good in transparency and robustness against various attacks such as cropping, Gaussian noise - adding, salt - pepper noise - adding, sharpening etc.
出处
《山东师范大学学报(自然科学版)》
CAS
2008年第2期14-17,共4页
Journal of Shandong Normal University(Natural Science)
基金
山东省教育厅科技计划项目(J05G20)
关键词
数字水印
近邻取样
彩色载体图像
Arnold置乱预处理
盲检测
digital watermarking
nearest neighbor algorithm
color host image
arnold shuffle pretreatment
blind extraction