Digital image watermark algorithm based on compressive sensing
Liang Li · Modern Electronics Technique · 2012
The compressive sensing theory(CS),as a new generation of signal processing theory foundation,has been applied in varieties of areas.But CS theory has yet to be further developed in the digital watermark application.In this paper,a new algorithm of digital image watermark based on CS is proposed to embed watermark in the observation domain of the sparsed image.Gaussian stochastic matrix is selected to observe images sparsed by wavelet transform and realize watermark embedding and extracting in the observation domain.And then a Matching Pursuit algorithm is used to restore sparse signal.The experimental results indicate that the algorithm meets the needs of transparency,robustness and security.At the same time,the anti-extraction experiments show that the algorithm can resist illegal extraction of the watermark efficiently,and the security is guaranteed.