Anti-forensics of double JPEG compression detection

Patchara Sutthiwan, Yun Q. Shi

Research output: Chapter in Book/Report/Conference proceedingConference contribution

24 Scopus citations


In this paper, a simple yet effective anti-forensic scheme capable of misleading double JPEG compression detection techniques is proposed. Based on image resizing with bilinear interpolation, the proposed operation aims at destroying JPEG grid structure while preserving reasonably good image quality. Given a doubly compressed image, our attack modifies the image by JPEG decompressing, shrinking and zooming the image with bilinear interpolation before JPEG compression with the same quality factor as used in the given image. The efficacy of the proposed scheme has been evaluated on two prominent double JPEG detection techniques and the outcome reveals that the proposed scheme is mostly effective, especially in the cases that the first quality factor is lower than the second quality factor.

Original languageEnglish (US)
Title of host publicationDigital-Forensics and Watermarking - 10th International Workshop, IWDW 2011, Revised Selected Papers
Number of pages14
StatePublished - 2012
Event10th International Workshop on Digital-Forensics and Watermarking, IWDW 2011 - Atlantic City, NJ, United States
Duration: Oct 23 2011Oct 26 2011

Publication series

NameLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume7128 LNCS
ISSN (Print)0302-9743
ISSN (Electronic)1611-3349


Other10th International Workshop on Digital-Forensics and Watermarking, IWDW 2011
Country/TerritoryUnited States
CityAtlantic City, NJ

All Science Journal Classification (ASJC) codes

  • Theoretical Computer Science
  • General Computer Science


  • Anti-Forensics
  • Bilinear Interpolation
  • Counter-Forensics
  • Double JPEG Compression
  • Histogram Equalization
  • Image Resizing
  • Image Rotation
  • Low-Pass Filtering
  • Shrink-and-Zoom
  • Tampering Detection


Dive into the research topics of 'Anti-forensics of double JPEG compression detection'. Together they form a unique fingerprint.

Cite this