An embedding strategy on fusing multiple image features for data hiding in multiple images

Yang, Junxue; Liao, Xin

doi:10.1016/j.jvcir.2020.102822

Cited by 13 publications

(11 citation statements)

References 23 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…At the same time, as the embedding capacity increases, the PSNR tends to be stable. When k=18, compared with the algorithm proposed in [18] (IEC), the algorithm proposed in [25] (DSL), the algorithm proposed in [26] (DCW), and the algorithm proposed in [27] (FMI), the average PSNR of the algorithm in this paper is increased by 27.05%, 12.73%, 21.58%, and 9.46% respectively, which shows that the algorithm in this paper has high invisibility.…”

Section: ) Analysis Of Perceptual Tamperingmentioning

confidence: 83%

“…The experimental environment is Matlab7.0, Python 3.4.6, and the carrier set comes from Corel-5K, GHIM-20, Caltech 101. The algorithms proposed in [18], [25], [26], and [27] are selected as the experimental references.…”

Section: B Experimental Comparisonsmentioning

confidence: 99%

“…Content may change prior to final publication. The proposed algorithm is compared with the algorithms proposed in [18], [25], [26], and [27]. The invisibility comparison experiment results are shown in Fig.…”

Section: ) Analysis Of Perceptual Tamperingmentioning

confidence: 99%

“…Compared with the IEC algorithm in [18], the DSL algorithm in [25], the DCW algorithm in [26], and the FMI algorithm in [27], the perceptual tampering is a unique characteristic of this algorithm. When a secret-containing image carrier is transmitted on an open channel, a third party may likely attempt to influence or damage the secretcontaining behavior.…”

Section: ) Perceptual Tampering Experimentsmentioning

confidence: 99%

“…These two strategies have highsecurity performance in multi-image information hiding. Yang et al [18] proposed a multi-image information hiding strategy by fusing multiple image features. The algorithm used multiple features to quantitatively describe the complexity of the image.…”

Section: Introductionmentioning

confidence: 99%

See 4 more Smart Citations

Identifiable Tampering Multi-Carrier Image Information Hiding Algorithm Based on Compressed Sensing

et al. 2020

View full text Add to dashboard Cite

Aiming at the security problem of secret information preprocessing and the difficulty of improving the capacity and robustness of the single-carrier image information hiding algorithm, an identifiable tampering multi-carrier image information hiding algorithm based on compressed sensing is proposed. Firstly, the angle structure descriptor feature vector was used to preprocess and classify the image carrier set. Secondly, the GHM multiwavelet transform was applied to different types of image carriers to obtain the secret information hiding area which can balance the invisibility and robustness. Thirdly, the secret image was processed by compressed sensing, the resulting observation matrix was decomposed by singular value, and the chaotic scrambling was encoded by logistic mapping. Finally, the secret information was embedded in the image singular value to complete the information hiding of different types of multiquantity image carriers. Combined with the angle structure descriptor of the image, the algorithm proposed an effective way to organize multiple carriers, which improved the embedding quality and efficiency of secret information. The verification data and segmented secret information classification and embedding strategy made the proposed algorithm have a keen ability to detect tampering and effectively improve the efficiency and integrity of secret information extraction. Experimental results show that compared with image sharing information hiding algorithm and the single-carrier information hiding algorithm based on compressed sensing, the invisibility and robustness of our algorithm are significantly improved. At the same time, the proposed algorithm has strong anti-analysis ability, can effectively resist most image processing attacks, and is suitable for large capacity secret communication and high-security applications. INDEX TERMS Multi-carrier information hiding, compressed sensing, angle structure descriptor, multiwavelet transform, singular value decomposition.

show abstract