基于RS纠错和分块DCT变换的改进LSB数字水印算法

An Improved LSB Digital Watermarking Algorithm Based on the RS Correction and DCT Domain

本文提出了一种基于DCT变换域的用改进LSB算法嵌入带RS纠错码的数字水印方法。该方法无需原始图像信息就可以提取数字图像中的嵌入数据,用来进行数字取证,确认多媒体内容的作者,鉴定多媒体数据的真实性,或提供相关时间、地点等信息。本算法把信息均匀地分布在了整个图像的数据中,增加了数据副本对有意义的嵌入信息进行恢复,并且由于在编码阶段增加了纠错码字,所以进一步提高了算法的鲁棒性。仿真实验结果表明,本文算法对噪声叠加、平滑滤波、剪切、旋转等图像处理操作具有较好的鲁棒性。

The article proposes an improved LSB watermarking algorithm based on the DCT domain and the RS correction. The algorithm can be used to get the embedded data without the original image information and also can be used to offer digital proofs, confirm the author, identify the authenticity of multimedia data, and offer the information of time or position. The algorithm distributes all data over the picture averagely, increases the backup message to restore the meaning of the embedded information and adds the correction message in the encoding process to increase the stability and robustness. The experimental result shows that the algorithm is robust to the common digital processing operations such as adding noise, smooth filtering, cutting, rotation,and so on.