基于DCT与分形理论的遥感影像鲁棒水印算法

Robust Watermarking Algorithm for Remote Sensing Image Based on DCT and Fractal

提出一种基于DCT和分形理论的遥感影像数字水印算法。首先利用分形技术确定自相似块作为水印嵌入区域;然后对水印图像进行Arnold置乱预处理,并将其扫描为一维序列;最后,对所选自相似块进行DCT变换,并将水印嵌入其中频系数。实验证明本文算法易于计算和实现,算法实现盲检测且有效提高了水印的不可见性,对JEPG压缩、滤波和叠加噪声等攻击具有较好的鲁棒性。

This paper proposes a meaningful digital watermarking algorithm for remote sensing image based on DCT and fractal theory. Firstly, it determines self-similar blocks for watermark embedding by using fractal technology, then scrambles the watermark with Arnold transformation and the scrambled watermark image next be scanned into one-dimensional sequence, at last it applies DCT to the selected self-similar blocks and embedds the watermark into the blocks. Experiments have shown that the algorithm proposed in this paper has good invisibility, simple calculation, easy realization, and extracting watermark without the original remote sensing image. And it is well robust against JEPG compression, filtering and noise superimposed attacking.