Analisis dan Simulasi Video Watermarking Menggunakan Metode Dual Tree Complex Wavelet Transform (DT-CWT) dan Singular Value Decomposition (SVD)

Arina Fadhilah, Bambang Hidayat, Ratri Dwi Atmaja


Video piracy is the act of obtaining, copying, and selling or distributing videos that already had the copyright without the consent of the copyright owner. Watermarking is a process which embeds an additional information in the host video signal so that the embedded watermark cannot be seen and difficult to be erased or altered. Video watermarking in this journal used a mp4 format video and two different watermark images. Host video frames are divided into two equal lots, some of the frames are embedded by watermark image 1, and the others are embedded by watermark images 2. The methods used are Dual-Tree Complex Wavelet Transform (DTCWT) and Singular Value Decomposition (SVD). The two watermarks are embedded and extracted in each subband at a depth level 1 to level 4 DTCWT - SVD with the aim of seeking the best subband and the best level for embedding and extracting. In the extraction testing, watermarked video is given several attacks before extraction process. Based on the MOS and PSNR value the DTCWT-SVD level for embedding process is level 4, and based on the MOS and MSE value, the best extraction images are produced from the level 3. The best subband for embedding watermark are the subbands with three parts such as {1,5}{1,1}{1,2} and {1,5}{1,2}{1,2}.


video watermarking, DTCWT, SVD

Full Text:




  • There are currently no refbacks.

View My Stats