Semi-fragile Watermarking For Video Content Authentication

With the development of multimedia network technology, people can acquire and disseminate video information more conveniently. But it also brings the issues about video copyright protection, content authenticity and integrity of information. Now people can easily edit and modify video through computer software, so the authenticity of video is doubt. Therefore, the research of video content authentication technology is necessary. And the research is popular in recent years. The digital watermarking technology is an effective method. It has many application fields and important theoretical significance, such as video authentication, copyright protection, broadcast monitoring and user tracking.Now the videos are often tampered maliciously, the existing video content integrity authentication methods were summarized and analyzed. The semi-fragile video watermarking algorithm can locate tampered region accurately and reconstruct the region. Finally, the proposed algorithm is verified through the adequate experiment. The main content and research work includes:1. Aiming at the problem that other algorithms can’t locate region accurately, I-frame was divided into two different areas for feature extraction and watermark embedding respectively. It can avoid watermarking to influencing the feature. The energy of DCT coefficient and the edge can keep same, that can be treated as features for generating watermarking. The filtering processing and mathematical morphology operation are good to get a stable characteristic and reduce the possibility of false alarm. Two features are combined that makes up the disadvantages some other methods may miss detection. And it improves the recall.2. A semi-fragile watermarking algorithm was proposed that can restore the feature of video content. Using compressive sensing technology for generating watermarking reduces the amount of watermarking data and the impact on video quality. By calculating the average value of 4 × 4 macro-blocks, the stable characteristics of the video content are obtained. When quantifying, the small number of "1" will be selected, the generated features are sparse. Disrupting watermarking can ensure the security of the watermark and recover the content-feature.3. The semi-fragile watermarking algorithm is theoretical analyzed and the experimental effectiveness are validated. Experimental results show that the proposed semi-fragile watermarking algorithm has good invisibility, can resist compressing and other conventional processing. It can detect malicious tampering and locate the tampered area. The watermarking is generated through compressive sensing technology. It ensures features completed and reduces the amount of watermark data. It is good to reduce the impact on video quality. The main features of tampering area can be reconstructed and recovered, which is good to speculate the tampering motivation of attackers.