Performance analysis of 2 level DWT-SVD based non blind and blind video watermarking using range conversion method

Abstract

The recent availability of inexpensive digital recording and storage devices have created an environment to obtain, replicate and distribute digital content without any loss in quality. Hence copyright protection and authentication have become inevitable. Leading to a need, to develop an efficient scheme for video watermarking, involving 2-level discrete wavelet transform (DWT) and proficient decomposition technique of singular value decomposition (SVD). Performance parameters like processing time, peak signal to noise ratio (PSNR) and normalized correlation (NC) are compared for non-blind and blind watermarking techniques. Focusing the PSNR and NC values are obtained as an average value of 40 dB and 0.99 respectively, when subjected to formidable noise attacks such as geometrical, filtering, salt and pepper noises. Range conversion method improves PSNR for challenging attacks like salt and pepper, Gaussian and Gamma noises.