Development of imperceptible and robust video watermarking techniques in dwtsvd domain

Abstract

Video piracy is the process of procuring, duplicating and distributing the copyrighted video content without obtaining the concurrence from the concept owner. We are living a world where video piracy has become an incredible challenge in the areas of movie, video surveillance system and social media due to availability of high-speed internet and facilitated network. Various algorithms are on the avail for video security in terms of spatial domain techniques, wavelet domain techniques and transform domain techniques. The comparative studies of those domains reveal that the wavelet domain based algorithms entail high level of robustness owing according to its multi resolution features. The traditional algorithms are subjected to unconceivable challenges to maintain robustness and imperceptibility for various image processing and frame processing attacks. The image based watermarking methods are not suitable for video watermarking due to redundant frames as it contains the motion frames. Research confirms that DWT-SVD based algorithms are capable to withstand various attacks with high imperceptibility and robustness. Due to specific research gaps, various DWT-SVD based algorithms are on the rise in the recent past. In most of the algorithms, the whole watermark embedded into entire video. In some of the methods, whole watermark did embed in the selected frames, which can be simply recognized by dropping attack. These existing algorithms do not combat various attacks like noise addition, filtering, brighten, blur compression, frame averaging, swapping and dropping etc. They do address the authentication and integrity issues as it reduces the perceptual quality and robustness of video and watermark. In these existing methods, the redundancy of frames in the video newline