博弈论模型下基于运动矢量的视频隐写算法

针对目前大多数视频隐写算法不满足Kerckhoffs准则进行了研究,在博弈论隐写模型的基础上,提出了一种新的基于运动矢量修改的H.264视频隐写算法.该算法利用人眼视觉特性中对运动矢量的方向和速度特性敏感程度不同计算失真代价函数,再根据博弈论相关理论结合失真代价函数得到每个运动矢量的嵌入概率,实现了一种在理论上满足Kerckhoffs准则的视频隐写算法.实验结果表明,与同类型视频隐写算法相比,在满嵌时该算法的PSNR和SSIM的平均变化值分别降低了18.5％和12％,具有较好的安全性和不可感知性.

Video steganography algorithm based on motion vector in game theory model

Most of the current video steganography algorithms do not meet the kerckhoffs criterion. Based on the game theory steganography model, this paper proposed a new H. 264 video steganography algorithm based on motion vector modification. The algorithm calculated the distortion cost function according to the different sensitivity of human visual characteristics to the direction and speed characteristics of the motion vector, and then obtained the embedding probability of each motion vector according to the game theory and the distortion cost function, and realized a video steganography algorithm which meets the Kerckhoffs criterion in theory. Experimental results show that, compared with the same type of video steganography algorithm, the change values of PSNR and SSIM of this algorithm are reduced by 18.9% and 5.9% respectively when it is fully embedded. This algorithm has better security and imperceptibility.