| Video robust digital watermarking is a technology that embeds a specific digital signal into video carrier for copyright protection,covert communications,decoding assistance,etc.In the existing research efforts,the influence of image complexity variation on watermarking has been ignored that leads to the embedding failure for frames with low complexity and the failure to make full use of redundancy for frames with high complexity.Therefore in this thesis,we research on video robust watermarking algorithms adaptive to frame complexity.The main contributions are as follows.(1)A watermarking algorithm adaptive to frame complexity is proposed.Repetition embedding is adopted.The complexity of current encoded key frame is predicted according to the total number of sub-macroblocks that meet the embedding conditions in previous key frame,and the repletion number is determined according to the predicted complexity.The experimental results using the codec JM19.0 based on H.264/AVC indicate that the proposed algorithm can achieve invisibility the stability of the code stream that outperforms the recent related efforts.It achieves acceptable robustness as well.(2)A digital watermarking algorithm that combines divided regions with error-correcting codes to embed watermarks is proposed.This algorithm first expands the watermark waiting to be embedded into a watermark with redundant bits according to the rule of 7-4 Hamming code,then mapping the watermark with redundant bits to different consecutive areas on the key frame,therefore,different bits can be embedded in different regions.The algorithm extracts the watermark according to the majority voting mechanism,and corrects the watermark after the extraction.The proposed algorithm is evaluated through simulation experiments using the codec JM19.0 based on H.264/AVC.The results indicate that the algorithm ensures good invisibility of the watermark,stability of the code stream,and better robustness compared to previous algorithms.Moreover,the proposed algorithm exhibits superior resistance to repeated compression compared to recent related literature.The algorithm can also solve the asynchronous problem of watermark extraction.(3)A software was developed for video watermarking.The software can generate watermark information with a single click based on the system time and has features for embedding and extracting watermarks from videos.Additionally,it allows playing videos that contain embedded watermarks.The previous algorithm is also implemented in this software.The research in this thesis addresses the issue that existing methods for video robust digital watermarking cannot fully embed digital watermarks in frames with low picture complexity,The region-based embedded watermarking algorithm combined with error-correcting coding proposed in this paper solves the asynchronous problem of extracting watermarks.The research in this paper is instructive to the research of digital watermarking based on repeated embedding of watermark bits. |
