Research On Large-capacity Reversible Image Watermarking Algorithm In Transform Domain

Reversible digital watermarking technology is widely used in military,medical and other high demanding areas in recent years,because it can completely restore the initial information of the carrier works.In this paper,we mainly study the large capacity reversible image watermarking algorithm in the transform domain.The work is as follows:A large capacity reversible watermarking algorithm based on generalized difference expansion and histogram translation is proposed.This algorithm improves the traditional generalized difference expansion method in transform domain,and makes the difference that can be used for embedding watermark information be fully utilized,and reduces the overall amplitude of the difference.It improves the watermark embedding capacity and the visual quality of the watermarked image.By improving the translation of difference histogram,we only shift the difference within the interval ?--TTTT)2,[),2[ to-),[ TT to reduce the number of translational difference,expand the space that can be used for watermark embedding,and enhance the embedding capacity.According to the experimental results,it can be seen that the proposed algorithm improves the embedding capacity of the watermark information while ensuring the better visual quality of the watermarked image.An optimal large capacity reversible watermarking algorithm based on prediction error and compression expansion is proposed.The algorithm combines the prediction error expansion and the compression expansion method in the transform domain.The larger difference is compressed into a smaller difference and then used to embed the watermark,which can not only expand the watermark embedding space,improves the embedding capacity,but also reduces the distortion of the carrier image caused by embedding watermark information,and improves the visual quality of the watermark image.At the same time,this algorithm introduces the idea of optimization.The particle swarm optimization algorithm based on differential evolution is applied to optimize the compression threshold T to get the optimal parameter T,which further reduces the embedding distortion and improves the visual quality of the watermarked image.According to the experimental results,it can be seen that the proposed algorithm can embed a larger volume of watermark information on the premise of getting better visual quality of the watermarked image.In conclusion,based on the research of reversible image watermarking technology based on transform domain,two large capacity reversible image watermarking algorithms based on transform domain are proposed in this paper.These two algorithms improve the embedding capacity of the watermark while reducing the embedding distortion.