A new video watermarking using redundant discrete wavelet in singular value decomposition domain with multi‐objective optimization

• Published in: Concurrency and computation Vol. 33; no. 13
• Main Authors: Kumar, Pala Mahesh, Srinivas, Kalyanapu, Prabhakar, Bitla
• Format: Journal Article
• Language: English
• Published: Hoboken, USA John Wiley & Sons, Inc 10-07-2021; Wiley Subscription Services, Inc
• Subjects: blind video watermarking; Digital imaging; Digital watermarking; Discrete Wavelet Transform; Domains; Image transmission; multi‐objective optimization; Optimization; Quality assessment; redundant discrete wavelet; Singular value decomposition; Video; wavelet transform; Wavelet transforms
• ISSN: 1532-0626; 1532-0634
• DOI: 10.1002/cpe.6217

Day by day, the transmission or sharing of digital information online is increasing due to the fact that Internet usage has become somewhat of an addiction for the majority of people. This results in larger copyright violation problems. Nowadays, most of the data or information is being transmitted in the form of digital images or videos, which are rather simple for the violators of copyright to forge or fake and then share for profit. Thus, digital watermarking came into existence as a possible solution to address these violations of copyright. This article proposes a new blind video watermarking (BVW) using redundant discrete wavelet transform (RDWT) in singular value decomposition (SVD) domain, which utilizes the advantages of both RDWT and SVD to embed and extract the watermark information into the cover video without degrading the quality of the watermark. In addition, an efficient meta‐heuristic optimization with multi‐objective optimization (MOO) is employed to further optimize the proposed RDWT‐SVD approach. Further, various attacks were applied to the watermarked frame to disclose the robustness of proposed BVW using RDWT‐SVD with MOO approach. Extensive simulations on several test videos with comparison to the conventional BVW methodologies exhibit the transcendency of the proposed BVW using RDWT‐SVD with MOO approach in terms of watermarking quality evaluation metrics such as peak signal‐to‐noise ratio (PSNR), structural similarity index, normalized correlation, and even that of root mean square error as well.