An Image Forgery Detection Solution based on DCT Coefficient Analysis

Hoai Nguyen, Florent Retraint, Frédéric Morain-Nicolier, Agnès Delahaies

2019

Abstract

JPEG compression and double JPEG compression introduces systematically some particular characteristics in the Discrete Cosine Transform (DCT) domain. In this paper, we propose a description of these characteristics. We also describe how to exploit these characteristics to introduce a new and efficient solution for estimating the quantization steps used in the first compression of a double-compressed image. We also introduce a method for detecting forgery in compressed images. Rapid, having a simple implementation and there is no need for training to be functional, the proposed solution gives, however, a performance auspicious. Its performance is demonstrated on simulated images and images retrieved from several public databases.

Paper Citation

in Harvard Style

Nguyen H., Retraint F., Morain-Nicolier F. and Delahaies A. (2019). An Image Forgery Detection Solution based on DCT Coefficient Analysis.In Proceedings of the 5th International Conference on Information Systems Security and Privacy - Volume 1: ICISSP, ISBN 978-989-758-359-9, pages 487-494. DOI: 10.5220/0007412804870494

in Bibtex Style

@conference{icissp19,
author={Hoai Nguyen and Florent Retraint and Frédéric Morain-Nicolier and Agnès Delahaies},
title={An Image Forgery Detection Solution based on DCT Coefficient Analysis},
booktitle={Proceedings of the 5th International Conference on Information Systems Security and Privacy - Volume 1: ICISSP,},
year={2019},
pages={487-494},
publisher={SciTePress},
organization={INSTICC},
doi={10.5220/0007412804870494},
isbn={978-989-758-359-9},
}

in EndNote Style

TY - CONF

JO - Proceedings of the 5th International Conference on Information Systems Security and Privacy - Volume 1: ICISSP,
TI - An Image Forgery Detection Solution based on DCT Coefficient Analysis
SN - 978-989-758-359-9
AU - Nguyen H.
AU - Retraint F.
AU - Morain-Nicolier F.
AU - Delahaies A.
PY - 2019
SP - 487
EP - 494
DO - 10.5220/0007412804870494