pith. machine review for the scientific record. sign in

arxiv: 2603.17679 · v2 · submitted 2026-03-18 · 💻 cs.CV

Recognition: 2 theorem links

· Lean Theorem

Illumination-Aware Contactless Fingerprint Spoof Detection via Paired Flash-Non-Flash Imaging

Roja Sahoo , Anoop Namboodiri
Authors on Pith no claims yet

Pith reviewed 2026-05-15 09:39 UTC · model grok-4.3

classification 💻 cs.CV
keywords contactless fingerprintspoof detectionpresentation attack detectionflash illuminationpaired imagingbiometricsliveness detectionactive sensing
0
0 comments X

The pith

Paired flash-non-flash images distinguish real contactless fingerprints from printed, digital, and molded spoofs by accentuating material and structure differences.

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

Contactless fingerprint systems lack physical contact cues, making spoof detection harder with single-image methods that generalize poorly. This work tests paired flash and non-flash captures as a simple active sensing approach where the flash version highlights properties like ridge visibility, subsurface scattering, micro-geometry, and surface oils that differ between genuine skin and common attack materials. Interpretable metrics on lighting-induced differences, including inter-channel correlation, specular reflections, texture realism, and differential imaging, separate genuine samples from printed, digital, and molded fakes. The analysis covers preliminary results plus practical limits such as sensitivity to settings and dataset scale.

Core claim

Paired flash-non-flash contactless fingerprint acquisition serves as a lightweight active sensing mechanism for spoof detection, with flash illumination accentuating material- and structure-dependent properties including ridge visibility, subsurface scattering, micro-geometry, and surface oils while non-flash images supply baseline appearance context, allowing complementary interpretable metrics to discriminate genuine fingerprints from printed, digital, and molded presentation attacks.

What carries the argument

Paired flash-non-flash imaging as an active sensing mechanism that captures lighting-induced differences via metrics such as inter-channel correlation, specular reflection, texture realism, and differential imaging.

If this is right

  • The method improves generalization over single-image appearance features by adding physics-based cues from illumination differences.
  • Complementary metrics from the pair provide more interpretable decisions than black-box classifiers alone.
  • Limitations in imaging settings and dataset scale must be addressed to reach operational robustness.
  • The approach motivates physics-informed feature design for future contactless presentation attack detectors.

Where Pith is reading between the lines

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

  • Mobile devices could adopt brief paired captures to add liveness without extra hardware.
  • Combining the metrics with other modalities like depth or thermal data might further reduce vulnerability to advanced spoofs.
  • Automated flash intensity calibration per device could reduce sensitivity to capture settings.

Load-bearing premise

That the lighting-induced differences captured by the chosen metrics remain consistent and discriminative across devices, capture conditions, spoof materials, and emerging high-fidelity spoofs.

What would settle it

A high-fidelity spoof whose flash and non-flash image pair yields the same inter-channel correlation, specular reflection, and differential imaging values as a genuine fingerprint would falsify the discrimination claim.

Figures

Figures reproduced from arXiv: 2603.17679 by Anoop Namboodiri, Roja Sahoo.

Figure 1
Figure 1. Figure 1: Blockwise OCL maps for sample flash and non-flash fingerprint [PITH_FULL_IMAGE:figures/full_fig_p002_1.png] view at source ↗
Figure 2
Figure 2. Figure 2: LCS and ridge-valley intensity profiles for sample flash and non [PITH_FULL_IMAGE:figures/full_fig_p002_2.png] view at source ↗
Figure 3
Figure 3. Figure 3: Comparison of standard (AIT Sharpness, NFIQ2) and custom patch [PITH_FULL_IMAGE:figures/full_fig_p002_3.png] view at source ↗
Figure 5
Figure 5. Figure 5: Attention maps for sample flash and non-flash contactless fingerprint [PITH_FULL_IMAGE:figures/full_fig_p003_5.png] view at source ↗
Figure 6
Figure 6. Figure 6: Inter-channel correlation analysis for the sample dataset of flash and [PITH_FULL_IMAGE:figures/full_fig_p004_6.png] view at source ↗
Figure 7
Figure 7. Figure 7: Log-magnitude FFTs and radially averaged spectra are shown for (a) [PITH_FULL_IMAGE:figures/full_fig_p005_7.png] view at source ↗
read the original abstract

Contactless fingerprint recognition enables hygienic and convenient biometric authentication but poses new challenges for spoof detection due to the absence of physical contact and traditional liveness cues. Most existing methods rely on single-image acquisition and appearance-based features, which often generalize poorly across devices, capture conditions, and spoof materials. In this work, we study paired flash-non-flash contactless fingerprint acquisition as a lightweight active sensing mechanism for spoof detection. Through a preliminary empirical analysis, we show that flash illumination accentuates material- and structure-dependent properties, including ridge visibility, subsurface scattering, micro-geometry, and surface oils, while non-flash images provide a baseline appearance context. We analyze lighting-induced differences using interpretable metrics such as inter-channel correlation, specular reflection characteristics, texture realism, and differential imaging. These complementary features help discriminate genuine fingerprints from printed, digital, and molded presentation attacks. We further examine the limitations of paired acquisition, including sensitivity to imaging settings, dataset scale, and emerging high-fidelity spoofs. Our findings demonstrate the potential of illumination-aware analysis to improve robustness and interpretability in contactless fingerprint presentation attack detection, motivating future work on paired acquisition and physics-informed feature design. Code is available in the repository.

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.

Referee Report

2 major / 1 minor

Summary. The paper claims that paired flash-non-flash contactless fingerprint acquisition serves as a lightweight active sensing mechanism for spoof detection. Flash illumination accentuates material- and structure-dependent properties (ridge visibility, subsurface scattering, micro-geometry, surface oils) while non-flash images provide baseline context; interpretable metrics (inter-channel correlation, specular reflection, texture realism, differential imaging) are used to discriminate genuine fingerprints from printed, digital, and molded presentation attacks. The work is based on preliminary empirical analysis, examines limitations including sensitivity to imaging settings and emerging high-fidelity spoofs, and provides code.

Significance. If the central claim holds with proper validation, the approach offers a practical, interpretable, physics-informed alternative to single-image appearance-based methods for contactless fingerprint presentation attack detection, which is relevant for hygienic biometric systems. The explicit discussion of limitations and code release are strengths that support reproducibility and future work on paired acquisition.

major comments (2)
  1. Abstract: The central claim that the paired-imaging metrics reliably discriminate attacks rests on preliminary empirical analysis, but no quantitative results, error bars, sample counts per class, or dataset details are provided; this leaves the discriminative power and generalization un-supported for rigorous evaluation.
  2. Abstract: The load-bearing assumption that lighting-induced differences (ridge visibility, subsurface scattering, etc.) remain consistent across devices, capture conditions, spoof materials, and high-fidelity spoofs is stated but not validated; only printed, digital, and molded attacks are referenced without cross-device testing or scale information, undermining the lightweight active-sensing advantage.
minor comments (1)
  1. Abstract: Consider adding one or two concrete preliminary quantitative highlights (e.g., accuracy ranges or metric differences) to strengthen the summary even if full results appear later in the manuscript.

Simulated Author's Rebuttal

2 responses · 0 unresolved

We thank the referee for the constructive feedback and for recognizing the potential of paired flash-non-flash imaging as a lightweight active sensing approach. We address each major comment below and indicate the revisions we will make to strengthen the manuscript.

read point-by-point responses

  1. Referee: Abstract: The central claim that the paired-imaging metrics reliably discriminate attacks rests on preliminary empirical analysis, but no quantitative results, error bars, sample counts per class, or dataset details are provided; this leaves the discriminative power and generalization un-supported for rigorous evaluation.

    Authors: We agree that the abstract, as a concise summary, currently omits specific quantitative details. The full manuscript presents the preliminary empirical analysis with dataset descriptions and metric computations, but to make the central claim more transparent at the abstract level we will revise the abstract to include key quantitative indicators such as sample counts per class, discrimination performance figures, and reference to error bars shown in the figures. This revision will be limited to information already present in the manuscript. revision: yes

  2. Referee: Abstract: The load-bearing assumption that lighting-induced differences (ridge visibility, subsurface scattering, etc.) remain consistent across devices, capture conditions, spoof materials, and high-fidelity spoofs is stated but not validated; only printed, digital, and molded attacks are referenced without cross-device testing or scale information, undermining the lightweight active-sensing advantage.

    Authors: We acknowledge that the current experiments are confined to a single imaging setup and the three attack categories listed, without cross-device or high-fidelity spoof testing. The manuscript already contains an explicit limitations section discussing sensitivity to imaging settings and emerging spoofs. We will revise the abstract and the limitations discussion to state the tested scope more precisely, provide the exact dataset scale used, and clarify that consistency across devices remains an open question for future work. No new experiments will be added in this revision. revision: partial

Circularity Check

0 steps flagged

No circularity: empirical metrics derived directly from paired flash/non-flash images without self-referential definitions or fitted predictions

full rationale

The paper presents a preliminary empirical analysis of lighting-induced differences in contactless fingerprints using direct measurements such as inter-channel correlation, specular reflection, texture realism, and differential imaging. No equations, derivations, parameter fitting, or self-citations are described that reduce any claimed result to its own inputs by construction. The central claim rests on observable physical differences accentuated by flash illumination, which are treated as independent observations rather than outputs of a closed definitional loop. This is a standard non-circular empirical study whose validity depends on experimental scale and diversity, not on internal reduction to fitted parameters or prior author work.

Axiom & Free-Parameter Ledger

0 free parameters · 1 axioms · 0 invented entities

The work rests on standard domain assumptions about illumination effects on skin and materials, with no free parameters, invented entities, or ad-hoc axioms explicitly introduced in the provided abstract.

axioms (1)
  • domain assumption Flash illumination accentuates material- and structure-dependent properties including ridge visibility, subsurface scattering, micro-geometry, and surface oils
    Invoked in the preliminary empirical analysis section of the abstract as the basis for differential feature extraction.

pith-pipeline@v0.9.0 · 5514 in / 1260 out tokens · 28726 ms · 2026-05-15T09:39:17.217670+00:00 · methodology

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Lean theorems connected to this paper

Citations machine-checked in the Pith Canon. Every link opens the source theorem in the public Lean library.

What do these tags mean?
matches
The paper's claim is directly supported by a theorem in the formal canon.
supports
The theorem supports part of the paper's argument, but the paper may add assumptions or extra steps.
extends
The paper goes beyond the formal theorem; the theorem is a base layer rather than the whole result.
uses
The paper appears to rely on the theorem as machinery.
contradicts
The paper's claim conflicts with a theorem or certificate in the canon.
unclear
Pith found a possible connection, but the passage is too broad, indirect, or ambiguous to say the theorem truly supports the claim.

Reference graph

Works this paper leans on

33 extracted references · 33 canonical work pages · 2 internal anchors

  1. [1]

    Efficient face spoofing detection with flash,

    A. F. Ebihara, K. Sakurai, and H. Imaoka, “Efficient face spoofing detection with flash,”IEEE Transactions on Biometrics, Behavior, and Identity Science, vol. 3, no. 4, pp. 535–549, 2021

    work page 2021

  2. [2]

    Colfispoof: A new database for contactless fingerprint presentation attack detection research,

    J. Kolberg, J. Priesnitz, C. Rathgeb, and C. Busch, “Colfispoof: A new database for contactless fingerprint presentation attack detection research,” in2023 IEEE/CVF Winter Conference on Applications of Computer Vision Workshops (WACVW), 2023, pp. 653–661

    work page 2023

  3. [3]

    Fingerphoto spoofing in mobile devices: A preliminary study,

    A. Taneja, A. Tayal, A. Malhorta, A. Sankaran, M. Vatsa, and R. Singh, “Fingerphoto spoofing in mobile devices: A preliminary study,” in2016 IEEE 8th International Conference on Biometrics Theory, Applications and Systems (BTAS), 2016, pp. 1–7

    work page 2016

  4. [4]

    Presentation attack detection with advanced cnn models for noncontact-based fingerprint systems,

    S. Purnapatra, C. Miller-Lynch, S. Miner, Y . Liu, K. Bahmani, S. Dey, and S. Schuckers, “Presentation attack detection with advanced cnn models for noncontact-based fingerprint systems,” 2023. [Online]. Available: https://arxiv.org/abs/2303.05459

    work page arXiv 2023

  5. [5]

    On matching finger-selfies using deep scattering networks,

    A. Malhotra, A. Sankaran, M. Vatsa, and R. Singh, “On matching finger-selfies using deep scattering networks,”IEEE Transactions on Biometrics, Behavior, and Identity Science, vol. 2, no. 4, pp. 350–362, 2020

    work page 2020

  6. [6]

    Unconstrained fingerphoto database,

    S. Chopra, A. Malhotra, M. Vatsa, and R. Singh, “Unconstrained fingerphoto database,” in2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops (CVPRW), 2018, pp. 630– 6308

    work page 2018

  7. [7]

    Ridgebase: A cross-sensor multi-finger contactless fingerprint dataset,

    B. Jawade, D. D. Mohan, S. Setlur, N. Ratha, and V . Govindaraju, “Ridgebase: A cross-sensor multi-finger contactless fingerprint dataset,” in2022 IEEE International Joint Conference on Biometrics (IJCB). IEEE, Oct. 2022, p. 1–9

    work page 2022

  8. [8]

    Matching contactless and contact-based conven- tional fingerprint images for biometrics identification,

    C. Lin and A. Kumar, “Matching contactless and contact-based conven- tional fingerprint images for biometrics identification,”IEEE Transac- tions on Image Processing, vol. 27, no. 4, pp. 2008–2021, 2018

    work page 2008

  9. [9]

    Fusion2print: Deep flash-non-flash fusion for contactless fingerprint matching,

    R. Sahoo and A. Namboodiri, “Fusion2print: Deep flash-non-flash fusion for contactless fingerprint matching,” inProceedings of the 28th International Conference on Pattern Recognition (ICPR), Lyon, France, 2026

    work page 2026

  10. [10]

    Video-based fingerphoto recog- nition with anti-spoofing techniques with smartphone cameras,

    C. Stein, V . Bouatou, and C. Busch, “Video-based fingerphoto recog- nition with anti-spoofing techniques with smartphone cameras,” in 2013 International Conference of the BIOSIG Special Interest Group (BIOSIG), 2013, pp. 1–12

    work page 2013

  11. [11]

    RaspiReader: An Open Source Fingerprint Reader Facilitating Spoof Detection

    J. J. Engelsma, K. Cao, and A. K. Jain, “Raspireader: An open source fingerprint reader facilitating spoof detection,” 2017. [Online]. Available: https://arxiv.org/abs/1708.07887

    work page internal anchor Pith review Pith/arXiv arXiv 2017

  12. [12]

    A universal anti-spoofing approach for contactless fingerprint biometric systems,

    B. Adami, S. Tehranipoor, N. Nasrabadi, and N. Karimian, “A universal anti-spoofing approach for contactless fingerprint biometric systems,”

    work page

  13. [13]

    Available: https://arxiv.org/abs/2310.15044

    [Online]. Available: https://arxiv.org/abs/2310.15044

    work page arXiv

  14. [14]

    Gru-aunet: A domain adaptation framework for contactless fingerprint presentation attack detection,

    B. Adami and N. Karimian, “Gru-aunet: A domain adaptation framework for contactless fingerprint presentation attack detection,”

    work page

  15. [15]

    Available: https://arxiv.org/abs/2504.01213

    [Online]. Available: https://arxiv.org/abs/2504.01213

    work page arXiv

  16. [16]

    Presentation Attack Detection for Smartphone Based Fingerphoto Recognition Using Second Order Local Structures,

    P. Wasnik, R. Ramachandra, K. Raja, and C. Busch, “Presentation Attack Detection for Smartphone Based Fingerphoto Recognition Using Second Order Local Structures,” in2018 14th International Conference on Signal-Image Technology & Internet-Based Systems (SITIS), 2018, pp. 241–246

    work page 2018

  17. [17]

    Late deep fusion of color spaces to enhance finger photo presentation attack detection in smartphones,

    E. Marasco and A. Vurity, “Late deep fusion of color spaces to enhance finger photo presentation attack detection in smartphones,”Applied Sciences, vol. 12, no. 22, 2022

    work page 2022

  18. [18]

    Contactless fingerprint recognition and fingerprint spoof mitigation using cnn,

    K. Kinage, “Contactless fingerprint recognition and fingerprint spoof mitigation using cnn,”International Journal of Recent Technology and Engineering, vol. 8, 01 2020

    work page 2020

  19. [19]

    Contactless fingerprint biometric anti- spoofing: An unsupervised deep learning approach,

    B. Adami and N. Karimian, “Contactless fingerprint biometric anti- spoofing: An unsupervised deep learning approach,” 2023. [Online]. Available: https://arxiv.org/abs/2311.04148

    work page arXiv 2023

  20. [20]

    Clnet: a contactless fingerprint spoof detection using deep neural networks with a transfer learning approach,

    K. Rajaram, B. N.G., A. Gupthaet al., “Clnet: a contactless fingerprint spoof detection using deep neural networks with a transfer learning approach,”Multimedia Tools and Applications, vol. 83, pp. 27 703–27 722, 2024. [Online]. Available: https://doi.org/10.1007/ s11042-023-16511-6

    work page 2024

  21. [21]

    A spoof detection method for contactless fingerprint collection utilizing spectrum and polarization diversity,

    G. Abramovich, M. Ganesh, K. Harding, S. Manickam, J. Czechowski, X. Wang, and A. Vemury, “A spoof detection method for contactless fingerprint collection utilizing spectrum and polarization diversity,” in Proceedings of SPIE: Next-Generation Spectroscopic Technologies III, vol. 7680, April 2010, p. 768005

    work page 2010

  22. [22]

    Texture and wavelet-based spoof finger- print detection for fingerprint biometric systems,

    S. B. Nikam and S. Agarwal, “Texture and wavelet-based spoof finger- print detection for fingerprint biometric systems,” in2008 First Interna- tional Conference on Emerging Trends in Engineering and Technology, 2008, pp. 675–680

    work page 2008

  23. [23]

    Fingerprint quality and validity analysis,

    E. Lim, X. Jiang, and W. Yau, “Fingerprint quality and validity analysis,” inProceedings. International Conference on Image Processing, vol. 1, 2002, pp. I–I

    work page 2002

  24. [24]

    A comparative study of fingerprint image-quality estimation methods,

    F. Alonso-Fernandez, J. Fierrez, J. Ortega-Garcia, J. Gonzalez- Rodriguez, H. Fronthaler, K. Kollreider, and J. Bigun, “A comparative study of fingerprint image-quality estimation methods,” IEEE Transactions on Information Forensics and Security, vol. 2, no. 4. [Online]. Available: http://dx.doi.org/10.1109/TIFS.2007.908228

    work page doi:10.1109/tifs.2007.908228 2007

  25. [25]

    Nist fingerprint image quality 2,

    E. Tabassi, M. Olsen, O. Bausinger, C. Busch, A. Figlarz, G. Fiumara, O. Henniger, J. Merkle, T. Ruhland, C. Schiel, and M. Schwaiger, “Nist fingerprint image quality 2,” National Institute of Standards and Technology, Tech. Rep. 8382, 2021. [Online]. Available: https://doi.org/10.6028/NIST.IR.8382

    work page doi:10.6028/nist.ir.8382 2021

  26. [26]

    Towards using police officers’ business smartphones for contactless fingerprint acquisition and enabling fingerprint comparison against contact-based datasets,

    C. Kauba, D. S ¨ollinger, S. Kirchgasser, A. Weissenfeld, G. Fern ´andez Dom ´ınguez, B. Strobl, and A. Uhl, “Towards using police officers’ business smartphones for contactless fingerprint acquisition and enabling fingerprint comparison against contact-based datasets,”Sensors, vol. 21, no. 7, 2021

    work page 2021

  27. [27]

    Dinov2: Learning robust visual features without supervision,

    M. Oquab, T. Darcet, T. Moutakanni, H. V o, M. Szafraniec, V . Khalidov, P. Fernandez, D. Haziza, F. Massa, A. El-Nouby, M. Assran, N. Ballas, W. Galuba, R. Howes, P.-Y . Huang, S.-W. Li, I. Misra, M. Rabbat, V . Sharma, G. Synnaeve, H. Xu, H. Jegou, J. Mairal, P. Labatut, A. Joulin, and P. Bojanowski, “Dinov2: Learning robust visual features without supe...

    work page 2024

  28. [28]

    Deep Residual Learning for Image Recognition

    K. He, X. Zhang, S. Ren, and J. Sun, “Deep residual learning for image recognition,” 2015. [Online]. Available: https://arxiv.org/abs/1512.03385

    work page internal anchor Pith review Pith/arXiv arXiv 2015

  29. [29]

    Moir ´e attack (ma): A new potential risk of screen photos,

    D. Niu, R. Guo, and Y . Wang, “Moir ´e attack (ma): A new potential risk of screen photos,”arXiv preprint arXiv:2110.10444, 2021

    work page arXiv 2021

  30. [30]

    Tetrahedron based fast 3d fingerprint identifi- cation using colored leds illumination,

    C. Lin and A. Kumar, “Tetrahedron based fast 3d fingerprint identifi- cation using colored leds illumination,”IEEE Transactions on Pattern Analysis and Machine Intelligence, vol. 40, no. 12, pp. 3022–3033, 2018

    work page 2018

  31. [31]

    Local binary pattern(lbp) optimization for feature extraction,

    Z. Sedaghatjoo, H. Hosseinzadeh, and B. S. Bigham, “Local binary pattern(lbp) optimization for feature extraction,” 2024. [Online]. Available: https://arxiv.org/abs/2407.18665

    work page arXiv 2024

  32. [32]

    Grey level co-occurrence matrix (glcm) based second order statistics for image texture analysis,

    A. R. Zubair and O. A. Alo, “Grey level co-occurrence matrix (glcm) based second order statistics for image texture analysis,” 2024. [Online]. Available: https://arxiv.org/abs/2403.04038

    work page arXiv 2024

  33. [33]

    The optics of human skin,

    R. R. Anderson and J. A. Parrish, “The optics of human skin,”Journal of Investigative Dermatology, vol. 77, no. 1, pp. 13–19, 1981

    work page 1981