Authors
T.N.C. Doan, Florent Retraint, Cathel Zitzmann,
Title
Blind forensics tool of falsification for RAW images
In
2017 IEEE International Symposium on Signal Processing and Information Technology, ISSPIT 2017
Pages
18–23
Publisher
Institute of Electrical and Electronics Engineers Inc.
Year
2018
Publisher's URL
http://dx.doi.org/10.1109/ISSPIT.2017.8388312
Indexed by
Abstract
This paper presents a novel method for blind forgery detection of natural image in RAW format. The approach is based on a statistical noise model of natural RAW images. This model is characterized by two parameters which are used as a fingerprint to falsification identification. The identification is cast in the framework of the hypothesis testing theory. For practice use, the Generalized Likelihood Ratio Test (GLRT) is presented and its performance is theoretically established in case of unknown parameters where an estimation of those parameters is designed. Experiments with simulated and real images highlight the relevance of the proposed approach. © 2017 IEEE.
Affiliations
Offprint