¡¡Chinese Journal of Computers   Full Text
  TitleA Novel Secure Video Watermarking for Broadcast Monitoring
  AuthorsLIU Li1),2) PENG Dai-Yuan2)
  Address1)(Computer Science Application, Zhengzhou Institute of Aeronautical Industry Management, Zhengzhou 450015)
2)(School of Information Science & Technology, Southwest Jiaotong University, Chengdu 610031)
  Year2009
  IssueNo.11(2239¡ª2246)
  Abstract &
  Background 		Abstract Authentication of a video clip via its hash finds application in broadcast monitoring. For broadcast monitoring, a novel algorithm for computing a robust hash from video clips, being secret key based, is proposed based on the 3-D Discrete Wavelet Transform(3D-DWT) and median quantization. The proposed hash algorithm is shown to be remarkably robust against video modifications and lossy compression. On the basis of the video hashing, a secure video watermarking scheme for broadcast monitoring is put forward. In the proposed scheme, the short hash and copyright information are combined to create the embedding watermark. And they are used in the authentication of video clips. Some analyses and experimental results show that the proposed scheme has a good security and robustness.
Keywords video watermarking; video hashing; linear collusion; diversity Background
This work is supported by the National Natural Science Foundation of China(No.60572142), the Science & Technology of Henan province (No.0624220058) and the Science & Technology of the education department of Henan province of China (No.2007520045), etc.
In recent years, watermarking security has emerged as a new subject in the watermarking area. And it brings new challenges to the design of watermarking systems. This work describes one of the major problems for robust video watermarking in the application of broadcast monitoring, which is security. For broadcast monitoring, authentication of a video clip via its hash find application can solve the problem. At the same time, linear collusion is a critical issue in the current study of video watermarking, and the attack is traditionally in the watermarking literature concerned with robustness assessment. However, according to what Furon et al. said, linear collusion attack should be used to assess security.