A dual-tamper-detection method for digital image authentication and content self-recovery

Tong Liu, Xiaochen Yuan

Research output: Contribution to journalArticlepeer-review

5 Citations (Scopus)

Abstract

This paper proposes an approach to protect image content against malicious tampering based on watermarking technology. The watermark is composed of two kinds of check bits which are used for tampered region localization, and one recovery bit which is used for image recovery and is embedded into the three-Least Significant Bit planes of the original image. The first check bit is generated by applying the proposed Parity Check Bit Labeled method to each pixel, and the other is generated by employing hashing algorithm to each block after image decomposition. The superposition result detected from the two check bits contributes to lowering the probability of false-negative errors. Moreover, we propose a post-processing method Adaptive Structural Element Calculation which improves the accuracy of tamper detection result further. Experimental results show that our algorithm has good performance in keeping high quality of recovered image, and meanwhile improving the accuracy of tamper detection result.

Original languageEnglish
Pages (from-to)29805-29826
Number of pages22
JournalMultimedia Tools and Applications
Volume80
Issue number19
DOIs
Publication statusPublished - Aug 2021
Externally publishedYes

Keywords

  • Adaptive structural element calculation
  • Dual-tamper-detection
  • Parity check bit labeled
  • Tamper detection and content self-recovery

Fingerprint

Dive into the research topics of 'A dual-tamper-detection method for digital image authentication and content self-recovery'. Together they form a unique fingerprint.

Cite this