AI Summary & Key Takeaways
AI InsightThis novel system increases the robustness against geometric and temporal attacks, and increases the quality of the watermarked video.
📌 3 Key Takeaways
Abstract
Because of the large amount of frames, similarity between frames and temporal attacks (frame dropping, frame averaging, frame swapping etc.), the process for video watermarking is more difficult than that for image watermarking. Current image watermarking methods are limited to solve all these difficulties. We propose a novel video watermarking system based on the Hidden Markov Model (HMM). This novel watermarking scheme splits the video sequences into a Group of Pictures (GOP) with HMM. Portions of the binary watermark are embedded into each GOP with a wavelet domain watermarking algorithm. The embedding process is the standard additive algorithm [1] in low pass (LL) and high pass (HH) bands in the wavelet domain. This novel system increases the robustness against geometric and temporal attacks, and increases the quality of the watermarked video.
Özet
Because of the large amount of frames, similarity between frames and temporal attacks (frame dropping, frame averaging, frame swapping etc.), the process for video watermarking is more difficult than that for image watermarking. Current image watermarking methods are limited to solve all these difficulties. We propose a novel video watermarking system based on the Hidden Markov Model (HMM). This novel watermarking scheme splits the video sequences into a Group of Pictures (GOP) with HMM. Portions of the binary watermark are embedded into each GOP with a wavelet domain watermarking algorithm. The embedding process is the standard additive algorithm [1] in low pass (LL) and high pass (HH) bands in the wavelet domain. This novel system increases the robustness against geometric and temporal attacks, and increases the quality of the watermarked video.