ADAPTIVE EMBEDDING TOLERANCE BASED ON MOTION CHARACTERISTICS FOR IMPERCEPTIBLE VIDEO WATERMARKING

摘要

We describe a procedure that is based on the human visual system (HVS) to embed watermarking signals into digital video sequences in the wavelet domain. This procedure exploits the perceptual properties of HVS and the motion estimation of the content of video sequence to improve the imperceptibility of watermarking. Furthermore, the embedding tolerance of watermark could be adaptively calculated according to the motion characteristics of video. The experiment results indicate that this method can effectively reduce the degradation in the video quality.