【作者】 牛亚青； 张益；

【Author】 Niu Yaqing,Zhang Yi (School of Information Engineering,Communication University of China,Beijing 100024)

【机构】 中国传媒大学信息工程学院；

【摘要】 本文利用从视觉感知系统推导出的感知失真模型来控制视频图像的各个部分所能容忍的水印信号的最大强度,确定与图像相关的调制掩蔽系数,利用其来实现水印嵌入,从而避免对视频图像的感知效果下降,在达到视觉不可见性的条件下使嵌入强度更强。然而现有的基于视频图像内容的JND(Just Noticeable Distortion)模型都不尽完善,本文的技术关键点在于形成一个更完善的视频图像的感知失真控制模型,即一个更完善的JND模型。本文基于DCT变化域提出了一个结合时空灵敏度感知函数、眼球运动影响、亮度自适应、纹理掩蔽等多种因素影响的功能完善的JND模型。更多还原

【Abstract】 We present a novel invisible and robust watermarking technique for video data with the use of perceptual distortion control model in line with the Human Visual System(HVS) perception.The perceptual distortion control model can be translated into watermarking technology to determine video dependent upper bounds on watermark insertion. This allows us to provide the maximum strength transparent watermark.In this paper, we develop a more accurate perceptual distortion control model of video sequences,a DCT based JND model.This proposed model incorporates spatio-temporal contrast sensitivity function(CSF),the smooth pursuit eye movement(SPEM),the directionality of the motion in video sequences,luminance adaptation effect,and contrast masking effect based on block classification.更多还原