共 31 条
Reversible data hiding of JPEG images based on block sorting and segmented embedding
被引:5
作者:
Mao, Ningxiong
[1
]
He, Hongjie
[1
]
Chen, Fan
[2
]
Bellavista, Paolo
[3
]
Yang, Yaolin
[1
]
机构:
[1] Southwest Jiaotong Univ, Sch Informat Sci & Technol, Chengdu 611756, Peoples R China
[2] Southwest Jiaotong Univ, Sch Comp & Artificial Intelligence, Chengdu 611756, Peoples R China
[3] Univ Bologna, Sch Comp Sci & Engn, Viale Risorgimento 2, I-40136 Bologna, Italy
基金:
中国国家自然科学基金;
关键词:
JPEG image;
Reversible data hiding;
Block sorting;
Segmented embedding;
Dynamic allocation solution;
D O I:
10.1016/j.bspc.2023.105555
中图分类号:
R318 [生物医学工程];
学科分类号:
0831 ;
摘要:
Existing JPEG image reversible data hiding (RDH) schemes based on quantized DCT (Discrete cosine transform) modification usually use non-zero or zero AC (Alternating current) coefficients as peak points for data embed-ding. However, this approach tends to limit the embedding capacity or result in significant file size growth. This paper proposed an RDH scheme for JPEG images based on DCT block sorting and segmented embedding. This scheme can take into account the dual advantages of having a non-zero or zero AC coefficient as the peak point, so it has both high embedding capacity and low file growth, and good visual quality. Firstly, a new DCT block sorting method is designed, the smoothness of the block is estimated by the number of zero coefficients in the EOB (End of Block) and the fluctuation of the DC coefficient in the block, and the zero coefficients in the EOB in the block have not embedded data. Then, the sorted DCT block sequence is divided into three segments, the peak points of the front two segments are (-1, 1) and (-1, 0) respectively, and the third segment may be empty but it always remains unchanged. Finally, the dynamic allocation method is used to quickly solve the optimal seg-mentation point to achieve the optimal embedding performance. The experimental results show that the overall performance of this scheme is better than the existing state-of-the-art algorithms.
引用
收藏
页数:12
相关论文