Embedding and Extracting the Watermark in Video Files using Intelligent Techniques

Khaleel, Shahba I.

doi:10.33899/csmj.2011.163620

	Embedding and Extracting the Watermark in Video Files using Intelligent Techniques
AL-Rafidain Journal of Computer Sciences and Mathematics
Article 16, Volume 8, Issue 1, 2011, Pages 25-44 PDF (550.45 K)
Document Type: Research Paper
DOI: 10.33899/csmj.2011.163620
Author
Shahba I. Khaleel
College of Computer Science and Mathematics University of Mosul, Mosul, Iraq
Abstract
This research, design a watermarking system for embedding and extracting the watermark in video files. This research presented four-efficient and powerful ways to embed and extract the watermark, the extract which in turn characterized as blind since it does not need the original cover in the process of extracting the watermark. The first method established the mean and SVD, which relied on the conversion of Singular values decomposition SVD and calculate the average of the unique values which are resulting from the application of the SVD on each data frame of the video file that required for embedding by, and then this method has been improved by hybrid intelligence by using a genetic algorithm for embedding the watermark, which is a second method. And other two methods has been implemented in this research. The third and fourth methods use artificial neural networks to embed the watermark. Depending on the characteristics of the data frame we use elman and Jordan neural networks and we use genetic algorithm to generate the secret key. The four methods of watermark were efficient and robust against various attacks which was found by measuring the efficiency of the methods by calculating the values of the Peak Signal-to-Noise Ratio PSNR and the correlation coefficient Normalized Correlation Coefficient NCC.
Keywords
watermarking; genetic algorithm; artificial neural networks

Statistics Article View: 297 PDF Download: 263