【摘要】 提出一种新的盲水印算法,能从一个小图像子集上检测出水印,首先将图像进行小波变换,选择两个所需要的子带,利用相邻特征平均值和奇偶判决法在两个子带上各自重复内嵌一半水印,在水印检测过程中不需要原始图像。根据本算法的实验结果,该算法具有较好的透明性,对各种攻击有较强的顽健性,尤其是增强了小波域水印在面临裁剪攻击和联合攻击时的顽健性。更多还原

【Abstract】 A new blind watermark algorithm that can detect watermark from small image subset is proposed. Wavelet transform for the image is carried out and two required subsets are selected. Half a watermark will be embedded into two subsets by means of adjacent characteristic means and odd-even decision method, respectively. The original image is not needed during the watermark detecting process. The experimental results show that the algorithm has a better transparency and can resist various attacks. Especially the robustness in facing segmentation attack and joint attack is improved.更多还原