针对旋转、缩放和平移等几何攻击破坏水印检测同步，从而导致水印检测失败等问题，利用归一化图像的缩放、平移不变性及小波矩的旋转不变特性，结合奇偶量化技术，提出了一种有效抵抗几何攻击的数字图像水印新算法。首先利用图像归一化技术将原始载体映射到几何不变空间内；然后计算归一化图像的小波矩，并选取部分稳定的低阶小波矩用于水印嵌入；最后利用奇偶量化方法将水印信息嵌入到所选小波矩的不变量内。仿真实验结果表明，该图像水印方案不仅具有良好的透明性，而且具有较强的抵抗常规信号处理、几何攻击、联合攻击等能力。

In image watermarking, the watermark robustness to geometric transformations is still an open question. Using invariant image features to carry the watermark is an effective approach to address this problem. In this paper, a geometrically robust image watermarking algorithm using image normalization and wavelet moments is proposed. The geometrically invariant space is constructed by using image normalization. Some wavelet moments of the normalized image are selected, and their magnitudes are dither-modulated to embed the digital watermark. The watermarked image is obtained via reconstruction from the modified moments. Simulation results show that the hidden message can be decoded at low error rates, robust against image rotation, scaling and translation, and as well, a variety of other distortions such as lossy compression.