.png){kind=icon}
.png){kind=icon}
Return
Reversible image authentication with efficient block-wise tamper localization
.png){kind=icon}
PRE
AI
DOI：10.1016/j.jisa.2026.104463.png)
.png){kind=icon}
Original request for help.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
Reversible image authentication (RIA) schemes are systems based on reversible watermarking designed for image authentication. They can offer several advantages over other approaches, such as integrating the image and its authentication information into a single signal and providing block-wise tamper localization capability. However, existing RIA schemes report a large number of unembeddable blocks, and limited effectiveness in tamper localization and detection accuracy across different attacks. To handle these issues, this work proposes an RIA scheme with high-resolution tamper-localization capabilities and high detection accuracy. A reversible watermarking method is also proposed, which produces more embeddable blocks and less image distortion than existing techniques. Experimental results show that when a protected image is attacked, the tampered blocks are effectively detected and localized. In addition, compared to a state-of-the-art scheme, the proposed scheme is more effective for block-wise tamper detection against splicing, collage, and copy-move attacks, while it is competitive at the pixel level against these attacks.
Keywords:
Reversible image authentication
Reversible watermarking
Reversible data hiding
Tamper localization
Prediction error histogram shifting
Journal
IF:
3.7
Papers:
1.9K
Citations:
4.9K