Video Camera Identification from Sensor Pattern Noise with a Constrained ConvNet

The identification of source cameras from videos, though it is a highly relevant forensic analysis topic, has been studied much less than its counterpart that uses images. In this work we propose a method to identify the source camera of a video based on camera specific noise patterns that we extract from video frames. For the extraction of noise pattern features, we propose an extended version of a constrained convolutional layer capable of processing color inputs. Our system is designed to classify individual video frames which are in turn combined by a majority vote to identify the source camera. We evaluated this approach on the benchmark VISION data set consisting of 1539 videos from 28 different cameras. To the best of our knowledge, this is the first work that addresses the challenge of video camera identification on a device level. The experiments show that our approach is very promising, achieving up to 93.1% accuracy while being robust to the WhatsApp and YouTube compression techniques. This work is part of the EU-funded project 4NSEEK focused on forensics against child sexual abuse.

[1]  Roberto Caldelli,et al.  Multimedia Forensic Techniques for Acquisition Device Identification and Digital Image Authentication , 2010, Handbook of Research on Computational Forensics, Digital Crime, and Investigation.

[2]  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.

[3]  Jan Lukás,et al.  Detecting digital image forgeries using sensor pattern noise , 2006, Electronic Imaging.

[4]  Nasir D. Memon,et al.  Digital Single Lens Reflex Camera Identification From Traces of Sensor Dust , 2008, IEEE Transactions on Information Forensics and Security.

[5]  Belhassen Bayar,et al.  A Deep Learning Approach to Universal Image Manipulation Detection Using a New Convolutional Layer , 2016, IH&MMSec.

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

[7]  Zeno Geradts,et al.  Methods for identification of images acquired with digital cameras , 2001, SPIE Optics East.

[8]  Paolo Bestagini,et al.  An overview on video forensics , 2012, 2012 Proceedings of the 20th European Signal Processing Conference (EUSIPCO).

[9]  Paolo Bestagini,et al.  Demosaicing strategy identification via eigenalgorithms , 2014, 2014 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP).

[10]  Miroslav Goljan,et al.  Digital camera identification from sensor pattern noise , 2006, IEEE Transactions on Information Forensics and Security.

[11]  Belhassen Bayar,et al.  A Video Camera Model Identification System Using Deep Learning and Fusion , 2019, ICASSP 2019 - 2019 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP).

[12]  Marco Fontani,et al.  VISION: a video and image dataset for source identification , 2017, EURASIP Journal on Information Security.

[13]  Chang-Tsun Li,et al.  Source Camera Identification Using Enhanced Sensor Pattern Noise , 2009, IEEE Transactions on Information Forensics and Security.

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

[15]  George Azzopardi,et al.  Device-based Image Matching with Similarity Learning by Convolutional Neural Networks that Exploit the Underlying Camera Sensor Pattern Noise , 2020, ICPRAM.

[16]  Z. Jane Wang,et al.  Median Filtering Forensics Based on Convolutional Neural Networks , 2015, IEEE Signal Processing Letters.

[17]  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).

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

[19]  Marc Chaumont,et al.  Deep learning is a good steganalysis tool when embedding key is reused for different images, even if there is a cover sourcemismatch , 2015, Media Watermarking, Security, and Forensics.

[20]  Jiwu Huang,et al.  A universal image forensic strategy based on steganalytic model , 2014, IH&MMSec '14.