基于DCT域的AVS视频盲水印算法

结合人眼的视觉特性,提出了一种基于DCT域的AVS视频编解码标准的视频盲水印算法.算法首先对水印图像进行增强的Arnold变换,得到置乱的二值水印图像；然后根据嵌入公式替换嵌入点的系数,将水印信息自适应嵌入到Ⅰ帧的DCT系数中.提取水印时不需要原始视频,可实现盲水印.实验结果表明,该算法对高斯、帧剪切、盐噪声等攻击,具有较好的稳健性和不可见性.

A video blind watermark algorithm based on DCT domain AVS

Combined with human visual characteristics, a kind of blind watermarking algorithm in DCT domain of AVS Video codec standards is proposed. Firstly, the scrambled binary watermark image is presented after processing enhanced Arnold transform on the watermark image. And then, according to the embedded formula to replace the coefficient of the embedded point, watermark is adaptively embedded into the DCT coefficient of Frame I. Watermarking extraction can realize the blind watermarking without the original video. The experimental results show that the proposed algorithm has strong robustness and invisibility while attacking by the Gauss frame shear, salt noise and frame cutting.