An Anti-Forensic Algorithm of JPEG Double Compression Based Forgery Detection
Chunhui Feng, Zhengquan Xu, Zheng Xinghui · 2012
Nowadays, a number of forensic algorithms have been developed to detect the manipulations of digital images. However, little consideration has been given to anti-forensic schemes capable of fooling the forensic methods. Aiming at this situation, a novel anti-forensic algorithm is proposed to counter a forensic analysis that can detect double JPEG compression with the same quantization matrix, and the limitation of existing image forensics are illustrated accordingly. The proposed algorithm allows generating tampered images without overall recompression, therefore eliminates the double quantization artifacts which will generally exist in tampered JPEG images. Experiment results show that the tampered images generated by the presented method not only have a good disguise effect, but can also effectively counter the forgery detection method that is based on the detection of double JPEG compression with the same quantization matrix.