利用图像特征点及自相似性的鲁棒数字水印方案

Digital Image Watermarking Resistant to RST Attacks Based on Feature Points and Self-similarities within Images

利用图像潜在的自相似性并结合特征点提出了一个新的数字水印方案。首先提取图像最高3位位平面的重心确定特征三角形;然后,根据水印比特位的不同确定不同的寻找范围,对特征三角形寻找与其最相似的块,最后用最佳相似块来替换特征三角形,从而完成水印嵌入。在检测时用同样的方法提取特征三角形,并对特征三角形在整个图像范围内进行最佳相似块的寻找,根据相似块所在的区域来提取水印。实验表明,该方案对RST(旋转、放缩以及平移)、剪切等几何攻击具有很高的鲁棒性,对压缩、滤波和加噪声等常见的图像处理也取得了很好的效果。

A novel image watermarking scheme based on feature points and the self_similarities of the images is proposed. In the embedding, the gravity centers of the 3 most significant bit planes are calculated and construct a feature triangle. According to the watermark bit, we determine the searching area and look for the most similar triangle of the feature triangle. Then the feature triangle is replaced with the matching triangle and thus finish the embedding. In the detection, the feature triangle is obtained and the algortithm search for the most similar one within the whole image.Then the watermark bit can be decided according to the matching triangle position. Experimental results show that the scheme is robust to RST attacks as well as compression, filtering, cropping and additive noise.