Research on Algorithm of Image Reversible Watermarking Based on Compressed Sensing

摘要

To improve the robustness of existing reversible watermarking algorithm for digital images and enhance the imperceptibility of watermark images, a novel reversible watermarking algorithm based on compressed sensing theory was proposed. Merits of this algorithm were as follows: combined the compressed sensing theory with reversible watermarking technique based on image compression to compress the host image efficiently, so that the watermarking embedded effectively, the robustness and imperceptibility of watermarked image has been improved; use reconstruction algorithm of compressed sensing to reconstruct the cover images accurately so as to achieve reversible watermarking; the robustness and security of watermarking system is enhanced well by improved Arnold scrambling. Experimental results show that the proposed algorithm can not only achieve blind extraction, but also improve the robustness, imperceptibility and other performances significantly compared with similar algorithms. Moreover, it is easy to realize and the watermarked image has high quality.