Mislgan: An Anti-Forensic Camera Model Falsification Framework Using A Generative Adversarial Network

Deep learning techniques have become popular for performing camera model identification. To expose weaknesses in these methods, we propose a new anti-forensic framework that utilizes a generative adversarial network (GAN) to falsify an image's source camera model. Our proposed attack uses the generator trained in the GAN to produce an image that can fool a CNN-based camera model identification classifier. Moreover, our proposed attack will only introduce a minimal amount of distortion to the falsified image that is not perceptible to human eyes. By conducting experiments on a large amount of data, we show that the proposed attack can successfully fool a state-of-art camera model identification CNN classifier with 98% probability and maintain high image quality.

[1]  Yoshua Bengio,et al.  Generative Adversarial Nets , 2014, NIPS.

[2]  Christian Ledig,et al.  Photo-Realistic Single Image Super-Resolution Using a Generative Adversarial Network , 2016, 2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR).

[3]  K. J. Ray Liu,et al.  Temporal Forensics and Anti-Forensics for Motion Compensated Video , 2012, IEEE Transactions on Information Forensics and Security.

[4]  Hui Zeng,et al.  Countering anti-forensics of image resampling , 2015, 2015 IEEE International Conference on Image Processing (ICIP).

[5]  Matthew C. Stamm,et al.  Detecting anti-forensic attacks on demosaicing-based camera model identification , 2017, 2017 IEEE International Conference on Image Processing (ICIP).

[6]  Hany Farid,et al.  Digital Image Authentication From JPEG Headers , 2011, IEEE Transactions on Information Forensics and Security.

[7]  Paolo Bestagini,et al.  First Steps Toward Camera Model Identification With Convolutional Neural Networks , 2016, IEEE Signal Processing Letters.

[8]  Stefano Tubaro,et al.  Revealing the Traces of JPEG Compression Anti-Forensics , 2013, IEEE Transactions on Information Forensics and Security.

[9]  Nasir D. Memon,et al.  Source camera identification based on CFA interpolation , 2005, IEEE International Conference on Image Processing 2005.

[10]  Mauro Barni,et al.  Adversary-aware, data-driven detection of double JPEG compression: How to make counter-forensics harder , 2016, 2016 IEEE International Workshop on Information Forensics and Security (WIFS).

[11]  Min Wu,et al.  Nonintrusive component forensics of visual sensors using output images , 2007, IEEE Transactions on Information Forensics and Security.

[12]  Marc Chaumont,et al.  Camera model identification with the use of deep convolutional neural networks , 2016, 2016 IEEE International Workshop on Information Forensics and Security (WIFS).

[13]  Hui Zeng,et al.  Countering anti-forensics of median filtering , 2014, 2014 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP).

[14]  Monika Jain,et al.  Anti-forensic technique for median filtering using L1-L2 TV model , 2016, 2016 IEEE International Workshop on Information Forensics and Security (WIFS).

[15]  Matthew C. Stamm,et al.  Camera model identification framework using an ensemble of demosaicing features , 2015, 2015 IEEE International Workshop on Information Forensics and Security (WIFS).

[16]  Belhassen Bayar,et al.  Augmented convolutional feature maps for robust CNN-based camera model identification , 2017, 2017 IEEE International Conference on Image Processing (ICIP).

[17]  Matthew C. Stamm,et al.  Learned Forensic Source Similarity for Unknown Camera Models , 2018, 2018 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP).

[18]  Paolo Bestagini,et al.  A Counter-Forensic Method for CNN-Based Camera Model Identification , 2017, 2017 IEEE Conference on Computer Vision and Pattern Recognition Workshops (CVPRW).

[19]  Belhassen Bayar,et al.  Constrained Convolutional Neural Networks: A New Approach Towards General Purpose Image Manipulation Detection , 2018, IEEE Transactions on Information Forensics and Security.

[20]  Rob Fergus,et al.  Deep Generative Image Models using a Laplacian Pyramid of Adversarial Networks , 2015, NIPS.

[21]  Min Wu,et al.  Information Forensics: An Overview of the First Decade , 2013, IEEE Access.

[22]  Soumith Chintala,et al.  Unsupervised Representation Learning with Deep Convolutional Generative Adversarial Networks , 2015, ICLR.

[23]  K. J. Ray Liu,et al.  Anti-forensics of JPEG compression , 2010, 2010 IEEE International Conference on Acoustics, Speech and Signal Processing.

[24]  Heung-Kyu Lee,et al.  Median Filtered Image Restoration and Anti-Forensics Using Adversarial Networks , 2018, IEEE Signal Processing Letters.

[25]  Rainer Böhme,et al.  Hiding Traces of Resampling in Digital Images , 2008, IEEE Transactions on Information Forensics and Security.

[26]  Matthew C. Stamm,et al.  Computationally efficient demosaicing filter estimation for forensic camera model identification , 2016, 2016 IEEE International Conference on Image Processing (ICIP).

[27]  K. J. Ray Liu,et al.  Forensics vs. anti-forensics: A decision and game theoretic framework , 2012, 2012 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP).

[28]  Belhassen Bayar,et al.  Towards Open Set Camera Model Identification Using a Deep Learning Framework , 2018, 2018 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP).

[29]  Alex ChiChung Kot,et al.  Accurate Detection of Demosaicing Regularity for Digital Image Forensics , 2009, IEEE Transactions on Information Forensics and Security.

[30]  Florent Retraint,et al.  Camera Model Identification Based on the Heteroscedastic Noise Model , 2014, IEEE Transactions on Image Processing.

[31]  Rainer Böhme,et al.  The 'Dresden Image Database' for benchmarking digital image forensics , 2010, SAC '10.

[32]  Belhassen Bayar,et al.  Design Principles of Convolutional Neural Networks for Multimedia Forensics , 2017, Media Watermarking, Security, and Forensics.

[33]  K. J. Ray Liu,et al.  Anti-forensics of median filtering , 2013, 2013 IEEE International Conference on Acoustics, Speech and Signal Processing.

[34]  Mo Chen,et al.  Determining Image Origin and Integrity Using Sensor Noise , 2008, IEEE Transactions on Information Forensics and Security.

[35]  Min Wu,et al.  Robustness of Color Interpolation Identification against Anti-forensic Operations , 2012, Information Hiding.

[36]  Paolo Bestagini,et al.  Tampering Detection and Localization Through Clustering of Camera-Based CNN Features , 2017, 2017 IEEE Conference on Computer Vision and Pattern Recognition Workshops (CVPRW).

[37]  Mauro Barni,et al.  Hiding traces of median filtering in digital images , 2012, 2012 Proceedings of the 20th European Signal Processing Conference (EUSIPCO).