Skip to content

MKLab-ITI/image-forensics

 
 

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 
 
 
 
 
 
 
 
 

Repository files navigation

UPDATE (01/2019): We have now added a new Noise-based algorithm in the MATLAB toolbox (NOI5). Make sure to properly cite the original paper if you use it (see the corresponding README file).

UPDATE (07/2018): We have now added our novel CAGI tampering detection algorithm in both Matlab and Java. It is based on JPEG block grid inconsistencies, but also features significant post-processing to improve result quality. Feel free to try it out! You will find citation information in the README file in the coresponding Matlab folder.

image-forensics

This is an integrated framework for image forensic analysis. It includes a Java webservice, including seven splicing detection algorithm implementations, plus additional forensic tools, located in the subdirectory java_service and a Matlab algorithm evaluation framework, including implementations of a large number of splicing detection algorithms, located in the subdirectory matlab_toolbox. An demo providing online implementations of many of these algorithms can be found here: Online demo.

Citations

Please cite the following paper in your publications if you use the Java implementations:

@inproceedings{zamp16,
  author = "Markos Zampoglou and Symeon Papadopoulos and Yiannis Kompatsiaris and Ruben Bouwmeester and Jochen Spangenberg",
  booktitle = "Social Media In the NewsRoom, {#SMNews16@CWSM}, Tenth International AAAI Conference on Web and Social Media workshops",
  title = "Web and Social Media Image Forensics for News Professionals",
  year = "2016",
}

If you use the Matlab implementations, use the following citation:

@article{zampAcc,
  author = "Markos Zampoglou and Symeon Papadopoulos and Yiannis Kompatsiaris",
  title = "A Large-Scale Evaluation of Splicing Localization Algorithms for Web Images",
  journal = "Multimedia Tools and Applications",
  doi = "10.1007/s11042-016-3795-2"
  pages= "Accepted for publication",
}

In either case, you must also cite the original algorithm publication. The README file within each Matlab algorithm subfolder contains the corresponding citation.

Contact Markos Zampoglou [email protected] for any further information.

About

No description, website, or topics provided.

Resources

Stars

Watchers

Forks

Packages

No packages published

Languages

  • Java 63.7%
  • MATLAB 32.2%
  • C 3.2%
  • M 0.5%
  • Mathematica 0.2%
  • Objective-C 0.1%
  • HTML 0.1%