摘要
针对目前变换域脆弱水印算法对于篡改恢复能力的不足以及所用变换本身的局限性,提出了一种基于双树小波变换的图像编码自嵌入水印算法。算法的篡改定位和图像恢复分别对应着认证水印和恢复水印。算法以原始载体图像的DDWT域SPIHT编码压缩图像为恢复水印,将其嵌入到图像的小波分解低频系数上,以增强水印鲁棒性;以嵌入恢复水印的载体图像的低频系数大小特征为认证水印,将其嵌入在图像分块的最低有效位上。实验结果表明,本算法能准确定位非法的篡改操作并具有良好的自我恢复能力。
Currently, because of the limitations of fragile watermarking on tamper recovery and transform, this paper presented a self-embedding watermarking algorithm based on dual-tree wavelet Transform. Tamper localization and ima- ge recovery are corresponding to authentication watermarking and recovery watermarking. The recovery watermarking is spiht coding after carrier image DDWT transform, and it is embed in the low-frequency of wavelet transform. The au- thentication watermarking is the relationship of low-frequency coefficients after recovery watermarking is embed, and it is embed in the image sub-block of lsb. The experimental results show that the algorithm can accurately locate the illegal tamper and has nice self-recovery ability.
出处
《计算机科学》
CSCD
北大核心
2013年第8期282-284,288,共4页
Computer Science
