摘要
根据非抽样Contourlet变换的特性,提出一种非抽样Contourlet变换域的盲水印算法,算法选择在具有最大能量的中高频子带中嵌入水印,在对水印进行加密后,利用密钥生成PN序列,根据水印值的不同PN序列嵌入宿主图像的方式也不同。提取水印不需要原始图像,利用NSCT变换系数与PN序列的相关性系数提取水印,实现盲提取。实验证明算法具有一定的不可见性和鲁棒性,具有很强的抗JPEG压缩的鲁棒性。
According to the feature of the Nonsubsampled Contourlet transform,a blind digital watermarking is proposed.The algorithm chooses the sub-band which has the largest energy to embed the watermark.After encrypt the watermark,use the key to generate the PN sequence,the PN sequence will be embedded in the host image in different ways according to the different value of the watermark.Extract the watermark based on the correlation values of the PN sequence and the sub-band without the original image,achieved the blind extract.Results show that the algorithm has a certain degree of invisibility and robustness.Especially has great robustness against the JPEG compression.
出处
《计算机与数字工程》
2010年第10期113-115,193,共4页
Computer & Digital Engineering
基金
湖南省教育厅资助科研项目(编号:06C522)资助
