Detection and localization of forgery using statistics of DCT and Fourier components
作者：
Highlights：
• This paper presents an integrated technique for detection and localization of splicing as well as copy-move forgery in JPEG compressed images.
• The idea is to exploit doubly stochastic model of quantized discrete cosine transform (DCT) coefficients to detect forgery.
• Splicing localization is performed using block-wise correlation maps of dequantized DCT coefficients and its recompressed version at different quality factors.
• Copy-move forgery localization is performed by utilizing the principle moments of phase congruency.
• The proposed scheme is robust against various pre- and post-processing transformations.
摘要
•This paper presents an integrated technique for detection and localization of splicing as well as copy-move forgery in JPEG compressed images.•The idea is to exploit doubly stochastic model of quantized discrete cosine transform (DCT) coefficients to detect forgery.•Splicing localization is performed using block-wise correlation maps of dequantized DCT coefficients and its recompressed version at different quality factors.•Copy-move forgery localization is performed by utilizing the principle moments of phase congruency.•The proposed scheme is robust against various pre- and post-processing transformations.
论文关键词：Discrete cosine transform,Doubly stochastic model,Image forgery detection,JPEG compression,Phase congruency
论文评审过程：Received 5 July 2019, Revised 31 December 2019, Accepted 31 December 2019, Available online 7 January 2020, Version of Record 10 January 2020.
论文官网地址：https://doi.org/10.1016/j.image.2020.115778
