Algoritmo de marca de agua para video basado en el svh, robusto a múltiples conversiones de formato de video

This paper proposes a video watermarking algorithm based on image normalization, motion estimation and perceptual information performed in the baseband domain. The proposed algorithm is designed to survive legitimate operations and malicious attacks, focused on the robustness against multiple video format conversion. Watermark embedding and detection are carried out in the Discrete Cosine Transform domain, and image normalization is used to get geometric invariant feature of video frames. The watermark energy is adapted using perceptual information and motion vectors to get higher imperceptibility and robustness. Computer simulation results show the watermark robustness to the most common video codec conversions, geometric distortions, as well as common signal processing and intentional frame-based attacks.