Search

To improve visual quality and embedding rate of existing difference expansion reversible image watermarking algorithm, a reversible image watermarking algorithm was proposed based on bit plane decomposition. The carrier image was decomposed by bit plane, and the low 6 bit planes were combined together to form a new image. The new image was divided into non overlapping blocks. The entropy of each block was calculated, and the image blocks with small entropy were selected to embed into watermark by the generalized difference expansion algorithm. The overflow graph and the relevant auxiliary information were embedded into the selected original image blocks of the corresponding image blocks with large entropy by difference quantization method. The 8 bit gray scale images of Lena and Barbara were used to test the reversible watermarking algorithm. The results show that the algorithm has high embedding rate and high visual quality, and the original image can be fully restored. When Lena is used as host image, the payload reaches 0.93 bpp. Compared with other algorithms, the watermark embedding rate is improved with good visual quality. [ABSTRACT FROM AUTHOR]