基于奇异值分解与蜂群优化的鲁棒图像水印算法

Singular Value Decomposition and Bee Colony Optimization Based Robust Image Watermark Algorithm

针对图像水印算法对强力攻击鲁棒性较弱的问题,提出一种基于奇异值分解(SVD)与蜂群优化(ABC)的鲁棒水印算法。首先,采用重分布不变小波变换对原图像进行处理;然后,将小波变换域低频带分割为不重叠的若干块,使用人类视觉系统选择目标嵌入块;最终,修改SVD分解的左奇异向量矩阵第一列系数,将水印数据嵌入目标块。此外采用右奇异向量补偿水印嵌入引起的视觉失真,并采用ABC对嵌入阈值与补偿参数进行优化。实验结果表明,该方法对图像处理攻击具有强鲁棒性,同时具有较好的视觉质量。

Aimed at the problem that image watermark algorithm is not robust to hard attacks, a singular value decomposition（SVD） and bee colony based robust image watermark algorithm is proposed. Firstly, redistributed invariant wavelet transform is used to process the original image; then, the low-band of the wavelet transform image is divided into non-overlapping blocks, the human vision system is used to select the target block for embedding; lastly, the first column of the left singular vector matrix of SVD decomposition is merged for embedding the watermark data into the target block. At the same time, the right singular vector is used to compensate the vision distortion caused by watermark embedding, and ABC is used to optimize the embedding threshold and compensation parameters. Experiments results show that, the proposal is hard robust against image manipulation attack, and can obtain good vision quality.