基于图像不变特征抗几何攻击图像水印算法

Invariant Feature-based Image Watermarking Scheme Robust to Geometric Attacks

针对几何攻击容易使常规水印算法检测失败的问题,提出了一种基于图像不变特征的鲁棒数字水印算法。先利用SIFT检测器在归一化图像中提取图像特征点,然后在原图像中找到对应的不变特征区域,最后将水印信息重复嵌入到多个不重叠的局部特征区域的DWT系数中。实验结果表明,该算法对常规的信号处理攻击,旋转、缩放、剪切等攻击均具有很好的鲁棒性。

This paper discusses robust image watermarking method using invariant features.Firstly,the sift detector is adopted to extract features points from the normalization image,and then find the corresponding points are found in the original image.Finally,watermarking informations are embedded in the DWT coefficients of the non-overlapped local regions.Experimental results show that the proposed scheme is not only common signal processing attacks but also geometric attacks.