一种基于邻域均值关系的图像盲水印算法

Blind image watermarking algorithm based on neighboring mean value relationship

提出了一种基于邻域均值关系的静止图像盲水印算法。该算法在嵌入前对水印进行置乱处理以提高安全性,然后对原始图像进行二级离散小波变换,在变换后的低频子带中选择感知重要系数作为水印的嵌入区间,再根据邻近特征平均值,选择合适的嵌入位置,将水印重复嵌入到低频子带中。实验结果表明,该算法嵌入信息量大,水印透明性好,且对常见的滤波、JPEG压缩、添加噪声、剪切等多种攻击有较强的鲁棒性。

This paper puts forward a blind image watermarking algorithm based on neighboring mean value relationship. In this algorithm, the watermark is random disposed before embedded into the original image to improve security, then a 2-level DWT decomposition of the original image is performed and the perceptual significant coefficients are selected from LL sub-band to be the embedded domain. Finally, the watermark is embedded into the chosen positions repetitively according to the mean value of neighboring symbol. The experimental results show that this algorithm has the characteristics of large embedded capacity and good transparency, and is robust to the common image processing such as filter, JPEG compression, noising, image clipping and etc.