PRNU Based Source Camera Attribution for Image Sets Anonymized with Patch-Match Algorithm

Ahmet Emir Dirik; Ahmet Karak\"u\c{c}\"uk

arxiv: 1906.11871 · v1 · pith:VKVK3DFYnew · submitted 2019-06-27 · 📡 eess.IV · cs.MM· eess.SP

PRNU Based Source Camera Attribution for Image Sets Anonymized with Patch-Match Algorithm

Ahmet Karak\"u\c{c}\"uk , Ahmet Emir Dirik This is my paper

Pith reviewed 2026-05-25 13:50 UTC · model grok-4.3

classification 📡 eess.IV cs.MMeess.SP

keywords PRNUsource camera identificationPatch-Matchimage anonymizationdigital forensicsnoise patterncamera attributionstructural editing

0 comments

The pith

Randomized subsets of Patch-Match anonymized images retain enough residual PRNU signal for conventional source camera identification to succeed.

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The paper demonstrates that the Patch-Match algorithm, which disrupts the photo-response non-uniformity pattern to anonymize images at a reported 97 percent rate, does not fully eliminate the possibility of source attribution. By applying standard PRNU-based source camera identification to randomized subsets drawn from the edited images, the method recovers the originating camera even when the collection includes images from unknown devices. This approach is tested in two adversary scenarios where an attacker uses Patch-Match to alter incriminating photos taken with their own camera. The result provides a direct counter to the claim that Patch-Match leaves no recourse for PRNU forensics.

Core claim

The paper claims that randomized subsets of images processed by the Patch-Match algorithm still contain detectable residual PRNU noise patterns, allowing traditional correlation-based source camera identification to attribute the set to the correct camera even when unknown-camera images are intermixed.

What carries the argument

Randomized subset sampling combined with conventional PRNU correlation, which extracts residual noise patterns that survive Patch-Match structural editing.

If this is right

Sets of Patch-Match edited images can be linked back to their source camera.
Attribution remains possible when the image set contains photos from unknown cameras.
The method applies to adversary scenarios in which Patch-Match is used to distort PRNU patterns.
Patch-Match no longer provides complete anonymity against PRNU-based forensics.

Where Pith is reading between the lines

These are editorial extensions of the paper, not claims the author makes directly.

Subset sampling may serve as a general tactic against other partial anonymization techniques that leave scattered noise intact.
Investigators could apply the approach to image collections edited by similar content-aware tools beyond Patch-Match.
The residual signal strength likely depends on how many original pixels survive the matching process, suggesting parameter-specific testing.
Larger or more diverse unknown-camera mixtures could reduce performance, pointing to a need for robustness checks.

Load-bearing premise

Randomized subsets of the Patch-Match-edited images still contain enough untouched PRNU signal for standard identification methods to distinguish the true camera.

What would settle it

A test in which randomized subsets drawn from Patch-Match images produce correlation scores to the true camera that are no higher than scores to unrelated cameras, or where identification accuracy falls below reliable thresholds across multiple trials.

Figures

Figures reproduced from arXiv: 1906.11871 by Ahmet Emir Dirik, Ahmet Karak\"u\c{c}\"uk.

**Figure 2.** Figure 2: Image subsets One approach for the attribution subsets of PM-images to their source cameras would be to generate PRNU fingerprints for all possible combinations in a given image dataset (N images), which would consist of K = 2N −1 fingerprints. Assuming there were N = 30 images, a total of 1.07×109 fingerprints would be needed. On the other hand, if we were to reduce the number of combinations, K number c… view at source ↗

**Figure 1.** Figure 1: We would like to highlight that, out of the images we mentioned we would use for evaluations of PM-images, there were a few images (1 for A57, 3 for D90), which were still identifiable by the PRNU based SCI. At a first glance, inclusion of these images in our evaluations might be more realistic. However, such images should be simply filtered-out by running individual images through the conventional PRNU ba… view at source ↗

**Figure 3.** Figure 3: Distributions of PRNU similarity and the image quality of each image in the PM [PITH_FULL_IMAGE:figures/full_fig_p010_3.png] view at source ↗

**Figure 4.** Figure 4: In (a) non-matching case (PM-image subsets from Canon 60D and PRNU fingerprint [PITH_FULL_IMAGE:figures/full_fig_p012_4.png] view at source ↗

read the original abstract

Patch-Match is an efficient algorithm used for structural image editing and available as a tool on popular commercial photo-editing software. The tool allows users to insert or remove objects from photos using information from similar scene content. Recently, a modified version of this algorithm was proposed as a counter-measure against Photo-Response Non-Uniformity (PRNU) based Source Camera Identification (SCI). The algorithm can provide anonymity at a great rate (97\%) and impede PRNU based SCI without the need of any other information, hence leaving no-known recourse for the PRNU-based SCI. In this paper, we propose a method to identify sources of the Patch-Match-applied images by using randomized subsets of images and the traditional PRNU based SCI methods. We evaluate the proposed method on two forensics scenarios in which an adversary makes use of the Patch-Match algorithm and distorts the PRNU noise pattern in the incriminating images he took with his camera. Our results show that it is possible to link sets of Patch-Match-applied images back to their source camera even in the presence of images that come from unknown cameras. To our best knowledge, the proposed method represents the very first counter-measure against the usage of of Patch-Match in the digital forensics literature.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit. Tearing a paper down is the easy half of reading it; the pith above is the substance, this is the friction.

Desk Editor's Note private letter to a colleague

Randomized subsets offer a potential way around Patch-Match PRNU anonymization, but the lack of reported metrics leaves the practical effectiveness unclear.

read the letter

The punchline is that the authors present randomized subset selection as the first counter to Patch-Match anonymization for PRNU-based source camera identification. They show that by sampling subsets from the edited image set and applying conventional PRNU matching, it is possible to link the images back to the source camera in both single-camera and mixed-camera cases. What the paper does well is identify a gap: the Patch-Match method was proposed as leaving no recourse for PRNU SCI, and this work proposes a simple, label-free way to recover attribution anyway. It applies existing techniques to the new input without inventing new signal processing, which keeps the contribution focused. The soft spots are in the validation. The abstract states that results are positive on two scenarios but provides no numbers on accuracy, no description of the image sets used, and no details on subset size or selection process. This makes it difficult to assess whether the residual PRNU after the structural edits is sufficient for reliable detection, especially in the mixed-camera case where the method must isolate the correct source without prior labels. The concern that Patch-Match might suppress the signal too uniformly is reasonable and would need the full experiments to check. The work engages honestly with the literature on PRNU and the specific anonymization paper it counters. This paper is for the digital forensics community, particularly those working on camera source attribution and anti-forensics. A specialist reader would get value from the proposed countermeasure idea and could use it as a starting point for stronger experiments. It deserves peer review because the problem is timely and the approach is novel enough to warrant expert feedback on the method and results.

Referee Report

2 major / 0 minor

Summary. The manuscript presents a method for source camera attribution of image sets anonymized using the Patch-Match algorithm. By applying traditional PRNU-based source camera identification to randomized subsets of the images, the authors claim it is possible to link the sets back to their source camera even when mixed with images from unknown cameras. The approach is evaluated in two forensics scenarios where an adversary uses Patch-Match to distort the PRNU noise pattern, and the paper positions the work as the first counter-measure against Patch-Match in the digital forensics literature.

Significance. If the results hold, this would be significant as the first proposed counter-measure against Patch-Match anonymization in digital forensics. It demonstrates a creative extension of existing PRNU SCI techniques via randomized subsets and shows potential robustness of PRNU methods to structural editing that achieves high (97%) anonymity rates. The explicit positioning as a novel contribution and the focus on practical mixed-camera scenarios add value if supported by verifiable experiments.

major comments (2)

[Abstract] Abstract: the assertion of positive results on two scenarios supplies no quantitative metrics, dataset details, error rates, or methodological steps, making the central claim that randomized subsets retain detectable residual PRNU unverifiable. This is load-bearing for the contribution.
[Proposed Method] The method relies on the assumption that randomized subsets drawn from Patch-Match-anonymized images (claimed to achieve 97% anonymity by distorting PRNU) still contain enough residual PRNU signal for conventional correlation-based SCI to succeed. This assumption is least secure in the mixed-camera scenario, where subsets must isolate the source signal without labels; if structural edits suppress or decorrelate PRNU more uniformly than assumed, correlations would fall below decision thresholds even for true-source images. This is load-bearing for the central claim.

Simulated Author's Rebuttal

2 responses · 0 unresolved

We thank the referee for the constructive comments. We address each major point below and indicate where revisions will be made to strengthen the manuscript.

read point-by-point responses

Referee: [Abstract] Abstract: the assertion of positive results on two scenarios supplies no quantitative metrics, dataset details, error rates, or methodological steps, making the central claim that randomized subsets retain detectable residual PRNU unverifiable. This is load-bearing for the contribution.

Authors: We agree that the abstract, as currently written, is too high-level and does not supply the quantitative details needed to make the central claim immediately verifiable. In the revised version we will expand the abstract to include the key performance metrics (attribution success rates for each scenario), the sizes of the image sets and camera populations used, and the observed false-positive/false-negative rates, while remaining within the journal's length limit. revision: yes
Referee: [Proposed Method] The method relies on the assumption that randomized subsets drawn from Patch-Match-anonymized images (claimed to achieve 97% anonymity by distorting PRNU) still contain enough residual PRNU signal for conventional correlation-based SCI to succeed. This assumption is least secure in the mixed-camera scenario, where subsets must isolate the source signal without labels; if structural edits suppress or decorrelate PRNU more uniformly than assumed, correlations would fall below decision thresholds even for true-source images. This is load-bearing for the central claim.

Authors: The assumption is indeed central. Our experiments on the mixed-camera scenario demonstrate that, even though Patch-Match achieves 97 % anonymity under standard PRNU correlation, the residual signal remains sufficiently strong that repeated random subset sampling produces statistically higher correlation peaks for the true source camera than for unknown cameras. Because the subsets are drawn independently, the probability that a given subset contains enough unaltered PRNU pixels to exceed the decision threshold is non-zero; aggregating many such trials therefore isolates the source. We will add a dedicated paragraph in the method section explaining this statistical argument, together with an ablation study on subset size and the number of random trials, to make the robustness explicit. revision: partial

Circularity Check

0 steps flagged

No circularity: empirical application of standard PRNU SCI to new inputs

full rationale

The paper presents an empirical counter-measure that applies existing PRNU-based source camera identification techniques to randomized subsets of Patch-Match-anonymized images. No mathematical derivations, equations, parameter fits, or self-referential definitions appear in the provided text. The method is described as a direct use of traditional correlation-based SCI on new input data, with results evaluated experimentally on two forensics scenarios. No self-citation chains, ansatzes, or renamings reduce any claim to its own inputs by construction. The derivation chain is therefore self-contained against external benchmarks.

Axiom & Free-Parameter Ledger

0 free parameters · 0 axioms · 0 invented entities

Only the abstract is available; no free parameters, axioms, or invented entities are described at a level that can be extracted.

pith-pipeline@v0.9.0 · 5765 in / 1033 out tokens · 24584 ms · 2026-05-25T13:50:16.044381+00:00 · methodology

discussion (0)

Reference graph

Works this paper leans on

24 extracted references · 24 canonical work pages

[1]

Luk´ aˇ s, J

J. Luk´ aˇ s, J. Fridrich, M. Goljan, Determining digital image origin using sensor imperfections, in: A. Said, J. G. Apostolopoulos (Eds.), Proceedings of SPIE, Vol. 5685, SPIE, 2005, p. 249. doi:10.1117/12.587105. URL http://ieeexplore.ieee.org/lpdocs/epic03/wrapper.htm? arnumber=1530329https://proceedings.spiedigitallibrary.org/ proceeding.aspx?doi=10....

work page doi:10.1117/12.587105 2005
[3]

Luk´ aˇ s, J

J. Luk´ aˇ s, J. Fridrich, M. Goljan, Detecting digital image forgeries using sensor pattern noise, in: Proc.SPIE, Vol. 6072, 2006, pp. 6072 – 6072 – 11. doi:10.1117/12.640109. URL https://doi.org/10.1117/12.640109

work page doi:10.1117/12.640109 2006
[4]

Luk´ aˇ s, J

J. Luk´ aˇ s, J. Fridrich, M. Goljan, Digital Camera Identiﬁcation From Sensor Pattern Noise, IEEE Transactions on Information Forensics and Security 1 (2) (2006) 205–214. doi:10.1109/TIFS.2006.873602. URL http://ieeexplore.ieee.org/xpls/abs{_}all.jsp?arnumber= 1634362http://ieeexplore.ieee.org/lpdocs/epic03/wrapper.htm? arnumber=1634362http://ieeexplore....

work page doi:10.1109/tifs.2006.873602 2006
[5]

Goljan, M

M. Goljan, M. Chen, J. Fridrich, Identifying Common Source Digital Camera from Image Pairs, in: 2007 IEEE International Conference on Image Processing, no. 1, IEEE, 2007, pp. VI – 125–VI – 128. doi:10.1109/ICIP.2007.4379537. URL http://ieeexplore.ieee.org/lpdocs/epic03/wrapper.htm? arnumber=4379537http://ieeexplore.ieee.org/document/4379537/

work page doi:10.1109/icip.2007.4379537 2007
[6]

Goljan, J

M. Goljan, J. Fridrich, T. Filler, Large scale test of sensor ﬁngerprint camera identiﬁcation, in: E. J. Delp III, J. Dittmann, N. D. Memon, P. W. Wong (Eds.), Proceedings of SPIE, no. 607, International Society for Optics and Photonics, SPIE, 2009, p. 72540I. doi:10.1117/12.805701. URL http://proceedings.spiedigitallibrary.org/data/ Conferences/SPIEP/177...

work page doi:10.1117/12.805701 2009
[7]

Goljan, J

M. Goljan, J. Fridrich, Sensor-ﬁngerprint based identiﬁcation of images cor- rected for lens distortion, in: Media Watermarking, Security, and Forensics 2012, Vol. 8303, International Society for Optics and Photonics, 2012, p. 83030H. URL https://doi.org/10.1117/12.909659

work page doi:10.1117/12.909659 2012
[8]

T. Gloe, S. Pfennig, M. Kirchner, Unexpected artefacts in PRNU-based camera identiﬁcation, in: Proceedings of the on Multimedia and security - MM&Sec ’12, MM&Sec ’12, ACM Press, New York, New York, USA, 2012, p. 109. doi:10.1145/2361407.2361426. URL http://doi.acm.org/10.1145/2361407.2361426http://dl.acm. org/citation.cfm?doid=2361407.2361426

work page doi:10.1145/2361407.2361426 2012
[9]

Valsesia, G

D. Valsesia, G. Coluccia, T. Bianchi, E. Magli, User authentication via prnu-based physical unclonable functions, Trans. Info. For. Sec. 12 (8) (2017) 1941–1956. doi:10.1109/TIFS.2017.2697402

work page doi:10.1109/tifs.2017.2697402 2017
[10]

Goljan, J

M. Goljan, J. Fridrich, Sensor ﬁngerprint digests for fast camera iden- tiﬁcation from geometrically distorted images, in: A. M. Alattar, N. D. Memon, C. D. Heitzenrater (Eds.), Proc. SPIE 8665, Media Watermarking, Security, and Forensics 2013, 86650B (22 March 2013), Vol. 8665, 2013, p. 86650B. doi:10.1117/12.2003234. URL http://ws2.binghamton.edu/fridri...

work page doi:10.1117/12.2003234 2013
[11]

Bayram, H

S. Bayram, H. T. Sencar, N. Memon, Eﬃcient Sensor Fingerprint Match- ing Through Fingerprint Binarization, IEEE Transactions on Information Forensics and Security 7 (4) (2012) 1404–1413. doi:10.1109/TIFS.2012. 2192272. URL http://ieeexplore.ieee.org/document/6175945/ 19

work page doi:10.1109/tifs.2012 2012
[12]

Taspinar, H

S. Taspinar, H. T. Sencar, S. Bayram, N. Memon, Fast camera ﬁngerprint matching in very large databases, in: Proceedings - International Confer- ence on Image Processing, ICIP, Vol. 2017-Septe, IEEE, 2018, pp. 4088–

work page 2017
[13]

URL http://ieeexplore.ieee.org/document/8297051/

doi:10.1109/ICIP.2017.8297051. URL http://ieeexplore.ieee.org/document/8297051/

work page doi:10.1109/icip.2017.8297051 2017
[14]

Goljan, J

M. Goljan, J. Fridrich, M. Chen, Defending Against Fingerprint- Copy Attack in Sensor-Based Camera Identiﬁcation, IEEE Transac- tions on Information Forensics and Security 6 (1) (2011) 227–236. doi:10.1109/TIFS.2010.2099220. URL http://ieeexplore.ieee.org/lpdocs/epic03/wrapper.htm? arnumber=5667057http://ieeexplore.ieee.org/document/5667057/

work page doi:10.1109/tifs.2010.2099220 2011
[15]

A. E. Dirik, A. Karak¨ u¸ c¨ uk, Forensic use of photo response non-uniformity of imaging sensors and a counter method, Optics Express 22 (1) (2014)

work page 2014
[16]

URL https://www.osapublishing.org/oe/abstract.cfm?uri= oe-22-1-470

doi:10.1364/OE.22.000470. URL https://www.osapublishing.org/oe/abstract.cfm?uri= oe-22-1-470

work page doi:10.1364/oe.22.000470
[17]

Karak¨ u¸ c¨ uk, A

A. Karak¨ u¸ c¨ uk, A. E. Dirik, Adaptive photo-response non-uniformity noise removal against image source attribution, Digital Investigation 12 (2015) 66–76. doi:10.1016/j.diin.2015.01.017. URL http://dx.doi.org/10.1016/j.diin.2015.01.017https: //linkinghub.elsevier.com/retrieve/pii/S1742287615000183

work page doi:10.1016/j.diin.2015.01.017 2015
[18]

A. E. Dirik, H. T. Sencar, N. Memon, Analysis of Seam-Carving-Based Anonymization of Images Against PRNU Noise Pattern-Based Source Attribution, IEEE Transactions on Information Forensics and Security 9 (12) (2014) 2277–2290. doi:10.1109/TIFS.2014.2361200. URL http://ieeexplore.ieee.org/xpls/abs{_}all.jsp?arnumber= 6914598http://ieeexplore.ieee.org/lpdocs...

work page doi:10.1109/tifs.2014.2361200 2014
[19]

Karak¨ u¸ c¨ uk, A

A. Karak¨ u¸ c¨ uk, A. E. Dirik, H. T. Sencar, N. D. Memon, Recent advances in counter PRNU based source attribution and beyond, in: A. M. Alattar, N. D. Memon, C. D. Heitzenrater (Eds.), Proceedings of SPIE - The International Society for Optical Engineering, Vol. 9409, 2015, p. 94090N. doi:10.1117/12.2182458. URL http://proceedings.spiedigitallibrary.or...

work page doi:10.1117/12.2182458 2015
[20]

Taspinar, M

S. Taspinar, M. Mohanty, N. Memon, PRNU-Based Camera Attribution from Multiple Seam-Carved Images, IEEE Transactions on Information Forensics and Security 12 (12) (2017) 3065–3080. doi:10.1109/TIFS. 2017.2737961. URL http://ieeexplore.ieee.org/document/8006244/

work page doi:10.1109/tifs 2017
[21]

Entrieri, M

J. Entrieri, M. Kirchner, Patch-Based Desynchronization of Dig- ital Camera Sensor Fingerprints, in: Electronic Imaging, Media 20 Watermarking, Security, and Forensics, Vol. 2016, 2016, pp. 1–9. doi:10.2352/ISSN.2470-1173.2016.8.MWSF-087. URL http://www.ingentaconnect.com/content/10.2352/ISSN. 2470-1173.2016.8.MWSF-087http://ws.binghamton.edu/kirchner/ pa...

work page doi:10.2352/issn.2470-1173.2016.8.mwsf-087 2016
[22]

Korus, J

P. Korus, J. Huang, Multi-Scale Analysis Strategies in PRNU-Based Tam- pering Localization, IEEE Transactions on Information Forensics and Se- curity 12 (4) (2017) 809–824. doi:10.1109/TIFS.2016.2636089

work page doi:10.1109/tifs.2016.2636089 2017
[23]

Korus, J

P. Korus, J. Huang, Evaluation of random ﬁeld models in multi-modal unsupervised tampering localization, in: 2016 IEEE International Work- shop on Information Forensics and Security (WIFS), 2016, pp. 1–6. doi: 10.1109/WIFS.2016.7823898

work page doi:10.1109/wifs.2016.7823898 2016
[24]

M. Chen, J. Fridrich, M. Goljan, Digital imaging sensor identiﬁcation (further study), in: A. J. e. a. Woods (Ed.), Proc. SPIE 6505, Security, Steganography, and Watermarking of Multimedia Contents IX, 2007, p. 65050P. doi:10.1117/12.703370. URL http://spiedigitallibrary.org/proceeding.aspx?doi=10. 1117/12.703370http://proceedings.spiedigitallibrary.org/ ...

work page doi:10.1117/12.703370 2007
[25]

M. Chen, J. Fridrich, M. Goljan, J. Luk´ aˇ s, Source digital camcorder identiﬁcation using sensor photo response non-uniformity, in: Electronic Imaging, Media Watermarking, Security, and Forensics, Vol. 6505, 2007, p. 65051G. doi:10.1117/12.696519. URL http://dx.doi.org/10.1117/12.696519http://proceedings. spiedigitallibrary.org/proceeding.aspx?doi=10.11...

work page doi:10.1117/12.696519 2007

[1] [1]

Luk´ aˇ s, J

J. Luk´ aˇ s, J. Fridrich, M. Goljan, Determining digital image origin using sensor imperfections, in: A. Said, J. G. Apostolopoulos (Eds.), Proceedings of SPIE, Vol. 5685, SPIE, 2005, p. 249. doi:10.1117/12.587105. URL http://ieeexplore.ieee.org/lpdocs/epic03/wrapper.htm? arnumber=1530329https://proceedings.spiedigitallibrary.org/ proceeding.aspx?doi=10....

work page doi:10.1117/12.587105 2005

[2] [3]

Luk´ aˇ s, J

J. Luk´ aˇ s, J. Fridrich, M. Goljan, Detecting digital image forgeries using sensor pattern noise, in: Proc.SPIE, Vol. 6072, 2006, pp. 6072 – 6072 – 11. doi:10.1117/12.640109. URL https://doi.org/10.1117/12.640109

work page doi:10.1117/12.640109 2006

[3] [4]

Luk´ aˇ s, J

J. Luk´ aˇ s, J. Fridrich, M. Goljan, Digital Camera Identiﬁcation From Sensor Pattern Noise, IEEE Transactions on Information Forensics and Security 1 (2) (2006) 205–214. doi:10.1109/TIFS.2006.873602. URL http://ieeexplore.ieee.org/xpls/abs{_}all.jsp?arnumber= 1634362http://ieeexplore.ieee.org/lpdocs/epic03/wrapper.htm? arnumber=1634362http://ieeexplore....

work page doi:10.1109/tifs.2006.873602 2006

[4] [5]

Goljan, M

M. Goljan, M. Chen, J. Fridrich, Identifying Common Source Digital Camera from Image Pairs, in: 2007 IEEE International Conference on Image Processing, no. 1, IEEE, 2007, pp. VI – 125–VI – 128. doi:10.1109/ICIP.2007.4379537. URL http://ieeexplore.ieee.org/lpdocs/epic03/wrapper.htm? arnumber=4379537http://ieeexplore.ieee.org/document/4379537/

work page doi:10.1109/icip.2007.4379537 2007

[5] [6]

Goljan, J

M. Goljan, J. Fridrich, T. Filler, Large scale test of sensor ﬁngerprint camera identiﬁcation, in: E. J. Delp III, J. Dittmann, N. D. Memon, P. W. Wong (Eds.), Proceedings of SPIE, no. 607, International Society for Optics and Photonics, SPIE, 2009, p. 72540I. doi:10.1117/12.805701. URL http://proceedings.spiedigitallibrary.org/data/ Conferences/SPIEP/177...

work page doi:10.1117/12.805701 2009

[6] [7]

Goljan, J

M. Goljan, J. Fridrich, Sensor-ﬁngerprint based identiﬁcation of images cor- rected for lens distortion, in: Media Watermarking, Security, and Forensics 2012, Vol. 8303, International Society for Optics and Photonics, 2012, p. 83030H. URL https://doi.org/10.1117/12.909659

work page doi:10.1117/12.909659 2012

[7] [8]

T. Gloe, S. Pfennig, M. Kirchner, Unexpected artefacts in PRNU-based camera identiﬁcation, in: Proceedings of the on Multimedia and security - MM&Sec ’12, MM&Sec ’12, ACM Press, New York, New York, USA, 2012, p. 109. doi:10.1145/2361407.2361426. URL http://doi.acm.org/10.1145/2361407.2361426http://dl.acm. org/citation.cfm?doid=2361407.2361426

work page doi:10.1145/2361407.2361426 2012

[8] [9]

Valsesia, G

D. Valsesia, G. Coluccia, T. Bianchi, E. Magli, User authentication via prnu-based physical unclonable functions, Trans. Info. For. Sec. 12 (8) (2017) 1941–1956. doi:10.1109/TIFS.2017.2697402

work page doi:10.1109/tifs.2017.2697402 2017

[9] [10]

Goljan, J

M. Goljan, J. Fridrich, Sensor ﬁngerprint digests for fast camera iden- tiﬁcation from geometrically distorted images, in: A. M. Alattar, N. D. Memon, C. D. Heitzenrater (Eds.), Proc. SPIE 8665, Media Watermarking, Security, and Forensics 2013, 86650B (22 March 2013), Vol. 8665, 2013, p. 86650B. doi:10.1117/12.2003234. URL http://ws2.binghamton.edu/fridri...

work page doi:10.1117/12.2003234 2013

[10] [11]

Bayram, H

S. Bayram, H. T. Sencar, N. Memon, Eﬃcient Sensor Fingerprint Match- ing Through Fingerprint Binarization, IEEE Transactions on Information Forensics and Security 7 (4) (2012) 1404–1413. doi:10.1109/TIFS.2012. 2192272. URL http://ieeexplore.ieee.org/document/6175945/ 19

work page doi:10.1109/tifs.2012 2012

[11] [12]

Taspinar, H

S. Taspinar, H. T. Sencar, S. Bayram, N. Memon, Fast camera ﬁngerprint matching in very large databases, in: Proceedings - International Confer- ence on Image Processing, ICIP, Vol. 2017-Septe, IEEE, 2018, pp. 4088–

work page 2017

[12] [13]

URL http://ieeexplore.ieee.org/document/8297051/

doi:10.1109/ICIP.2017.8297051. URL http://ieeexplore.ieee.org/document/8297051/

work page doi:10.1109/icip.2017.8297051 2017

[13] [14]

Goljan, J

M. Goljan, J. Fridrich, M. Chen, Defending Against Fingerprint- Copy Attack in Sensor-Based Camera Identiﬁcation, IEEE Transac- tions on Information Forensics and Security 6 (1) (2011) 227–236. doi:10.1109/TIFS.2010.2099220. URL http://ieeexplore.ieee.org/lpdocs/epic03/wrapper.htm? arnumber=5667057http://ieeexplore.ieee.org/document/5667057/

work page doi:10.1109/tifs.2010.2099220 2011

[14] [15]

A. E. Dirik, A. Karak¨ u¸ c¨ uk, Forensic use of photo response non-uniformity of imaging sensors and a counter method, Optics Express 22 (1) (2014)

work page 2014

[15] [16]

URL https://www.osapublishing.org/oe/abstract.cfm?uri= oe-22-1-470

doi:10.1364/OE.22.000470. URL https://www.osapublishing.org/oe/abstract.cfm?uri= oe-22-1-470

work page doi:10.1364/oe.22.000470

[16] [17]

Karak¨ u¸ c¨ uk, A

A. Karak¨ u¸ c¨ uk, A. E. Dirik, Adaptive photo-response non-uniformity noise removal against image source attribution, Digital Investigation 12 (2015) 66–76. doi:10.1016/j.diin.2015.01.017. URL http://dx.doi.org/10.1016/j.diin.2015.01.017https: //linkinghub.elsevier.com/retrieve/pii/S1742287615000183

work page doi:10.1016/j.diin.2015.01.017 2015

[17] [18]

A. E. Dirik, H. T. Sencar, N. Memon, Analysis of Seam-Carving-Based Anonymization of Images Against PRNU Noise Pattern-Based Source Attribution, IEEE Transactions on Information Forensics and Security 9 (12) (2014) 2277–2290. doi:10.1109/TIFS.2014.2361200. URL http://ieeexplore.ieee.org/xpls/abs{_}all.jsp?arnumber= 6914598http://ieeexplore.ieee.org/lpdocs...

work page doi:10.1109/tifs.2014.2361200 2014

[18] [19]

Karak¨ u¸ c¨ uk, A

A. Karak¨ u¸ c¨ uk, A. E. Dirik, H. T. Sencar, N. D. Memon, Recent advances in counter PRNU based source attribution and beyond, in: A. M. Alattar, N. D. Memon, C. D. Heitzenrater (Eds.), Proceedings of SPIE - The International Society for Optical Engineering, Vol. 9409, 2015, p. 94090N. doi:10.1117/12.2182458. URL http://proceedings.spiedigitallibrary.or...

work page doi:10.1117/12.2182458 2015

[19] [20]

Taspinar, M

S. Taspinar, M. Mohanty, N. Memon, PRNU-Based Camera Attribution from Multiple Seam-Carved Images, IEEE Transactions on Information Forensics and Security 12 (12) (2017) 3065–3080. doi:10.1109/TIFS. 2017.2737961. URL http://ieeexplore.ieee.org/document/8006244/

work page doi:10.1109/tifs 2017

[20] [21]

Entrieri, M

J. Entrieri, M. Kirchner, Patch-Based Desynchronization of Dig- ital Camera Sensor Fingerprints, in: Electronic Imaging, Media 20 Watermarking, Security, and Forensics, Vol. 2016, 2016, pp. 1–9. doi:10.2352/ISSN.2470-1173.2016.8.MWSF-087. URL http://www.ingentaconnect.com/content/10.2352/ISSN. 2470-1173.2016.8.MWSF-087http://ws.binghamton.edu/kirchner/ pa...

work page doi:10.2352/issn.2470-1173.2016.8.mwsf-087 2016

[21] [22]

Korus, J

P. Korus, J. Huang, Multi-Scale Analysis Strategies in PRNU-Based Tam- pering Localization, IEEE Transactions on Information Forensics and Se- curity 12 (4) (2017) 809–824. doi:10.1109/TIFS.2016.2636089

work page doi:10.1109/tifs.2016.2636089 2017

[22] [23]

Korus, J

P. Korus, J. Huang, Evaluation of random ﬁeld models in multi-modal unsupervised tampering localization, in: 2016 IEEE International Work- shop on Information Forensics and Security (WIFS), 2016, pp. 1–6. doi: 10.1109/WIFS.2016.7823898

work page doi:10.1109/wifs.2016.7823898 2016

[23] [24]

M. Chen, J. Fridrich, M. Goljan, Digital imaging sensor identiﬁcation (further study), in: A. J. e. a. Woods (Ed.), Proc. SPIE 6505, Security, Steganography, and Watermarking of Multimedia Contents IX, 2007, p. 65050P. doi:10.1117/12.703370. URL http://spiedigitallibrary.org/proceeding.aspx?doi=10. 1117/12.703370http://proceedings.spiedigitallibrary.org/ ...

work page doi:10.1117/12.703370 2007

[24] [25]

M. Chen, J. Fridrich, M. Goljan, J. Luk´ aˇ s, Source digital camcorder identiﬁcation using sensor photo response non-uniformity, in: Electronic Imaging, Media Watermarking, Security, and Forensics, Vol. 6505, 2007, p. 65051G. doi:10.1117/12.696519. URL http://dx.doi.org/10.1117/12.696519http://proceedings. spiedigitallibrary.org/proceeding.aspx?doi=10.11...

work page doi:10.1117/12.696519 2007