基于局部Tchebichef矩的鲁棒图像水印算法

针对目前基于图像不变特征水印算法不能同时有效抵抗常规图像处理和几何攻击这一问题,提出了一种基于局部Tchebichef矩(LTMs)的图像水印新算法。首先,利用Harris-Laplace检测算子提取载体图像多尺度空间中的特征点,并通过特征选择策略获得稳定且分离的局部圆形特征区域;然后,结合主方向对齐,得到具有旋转、缩放和平移(RST)不变性的局部圆形特征区域;最后,计算局部特征区域的Tchebichef矩,采用量化调制Tchebichef低阶矩幅值将水印嵌入到局部特征区域中。实验结果表明,本文算法在获得很好的不可见性的同时,对常规图像处理、几何攻击及组合攻击具有较好的鲁棒性。

Robust image watermarking scheme using local feature regions and tchebichef moments

Most previous watermarking schemes with image invariant features perform poorly under common image processing operations as well as geometric distortions.A robust image watermarking scheme based on local Tchebichef moments is proposed.Harris-Laplace detector is first adopted to extract feature points.Then,the stable and non-overlapped circular feature regions are chosen via the proposed feature selection criteria.Finally,Tchebichef moments are computed over these feature regions and watermark is embedded in...