pith. sign in

arxiv: 2606.13046 · v1 · pith:VUZA6CV2new · submitted 2026-06-11 · 📡 eess.SP

Thermal characterisation by Scanning Photothermal Radiometry using a random undersampled measurement scheme

Florian Crouau (I2M-BX) , Alejandro Mateos-Canseco (I2M-BX) , J\'er\'emie Maire (I2M-BX) , Jean-Luc Battaglia (I2M-BX) , St\'ephane Chevalier (I2M-BX) This is my paper

Pith reviewed 2026-06-27 05:52 UTC · model grok-4.3

classification 📡 eess.SP
keywords scanning photothermal radiometryrandom undersamplingsparse signalsthermal characterizationcarbon fibre aluminium compositeweighted random samplingnon-destructive testing
0
0 comments X

The pith

Scanning photothermal radiometry measurements can be reduced by a factor of six using weighted random undersampling on sparse signals.

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

This paper shows how to speed up scanning photothermal radiometry by reducing the number of measurements needed. SPR is a non-destructive contactless method for mapping thermal properties at high spatial and temporal resolution, but full scans take a long time. The authors apply irregular sampling to sparse signals from a carbon-fibre-in-aluminium sample and combine it with a weighted random selection technique. This combination cuts the measurement count by a factor of six while keeping the reconstructed thermal properties accurate.

Core claim

By employing a random undersampled measurement scheme that uses irregular sampling on sparse signals together with a weighted random selection technique, the number of measurements required for scanning photothermal radiometry can be reduced by a factor of six on a carbon fibre aluminium matrix sample while still permitting accurate reconstruction of the thermal properties.

What carries the argument

Weighted random undersampling scheme for irregular sampling of sparse thermal signals.

If this is right

  • Scanning time for thermal characterization of the composite is reduced proportionally to the undersampling factor.
  • Accurate depth-resolved thermal properties remain obtainable from the reduced dataset due to signal sparsity.
  • Weighted random selection improves reconstruction quality beyond plain irregular sampling.
  • The method works specifically because the thermal response of this carbon fibre-aluminium matrix exhibits sufficient sparsity.

Where Pith is reading between the lines

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

  • The same undersampling approach could shorten acquisition times in other contactless thermal or spectroscopic scanning methods that produce sparse responses.
  • If sparsity holds across a wider range of composites, the technique would expand practical use of SPR beyond laboratory settings.
  • Coupling the reduced measurements with faster reconstruction algorithms might enable near-real-time thermal mapping.

Load-bearing premise

The thermal signals from the carbon fibre-aluminium sample are sparse enough that weighted random undersampling by a factor of six still permits accurate reconstruction of the thermal properties.

What would settle it

If the thermal diffusivity or conductivity maps reconstructed from the undersampled data deviate beyond experimental uncertainty from those obtained with full sampling on the identical sample, the reduction claim would not hold.

Figures

Figures reproduced from arXiv: 2606.13046 by Alejandro Mateos-Canseco (I2M-BX), Florian Crouau (I2M-BX), Jean-Luc Battaglia (I2M-BX), J\'er\'emie Maire (I2M-BX), St\'ephane Chevalier (I2M-BX).

Figure 1
Figure 1. Figure 1: Schematic of Scanning Photothermal Radiometry Setup. A laser at 1064 nm focuses a beam of [PITH_FULL_IMAGE:figures/full_fig_p002_1.png] view at source ↗
Figure 2
Figure 2. Figure 2: Sub-sampling scheme at multiple frequencies, consisting in 2 parts: A sub-sampled acquisition [PITH_FULL_IMAGE:figures/full_fig_p004_2.png] view at source ↗
Figure 3
Figure 3. Figure 3: Schematic of the sample a, and its response at 6 kHz in amplitude b and phase c (Reconstruction [PITH_FULL_IMAGE:figures/full_fig_p005_3.png] view at source ↗
Figure 4
Figure 4. Figure 4: Reconstruction results for a 40 × 40 image with the two methods for different sampling ratios at a laser power of 750 mW. The first line contains the samples used for reconstruction in the other two lines. These samples are taken randomly at a given ratio, from 5% to 20%, compared to the full image. They are also such that each increase of samples always includes the points from the previous image. For the… view at source ↗
Figure 5
Figure 5. Figure 5: Reconstruction results for a 40 × 40 image taken at 6 kHz with the 3 methods (in lines) for different laser powers (in columns) ranging from 150 mW to 600 mW at a random sampling ratio of 25%. is the same as this is determined by the sampling ratio, but as they both require slower acquisition speed the minimum achievable time will be greater than for images of amplitude. 3.3 Multiple frequencies scan As de… view at source ↗
Figure 6
Figure 6. Figure 6: Reconstruction results for a 40 × 40 image at multiple frequencies. the aluminium, the probability could be defined as the inverse of what is shown here, ie. be maximum when the gradient is minimum. 4 Conclusion and Perspectives SPR can largely be accelerated thanks to algorithmic reconstruction of images from a small subset of samples. Both RBF interpolation and CS have been used to accurately represent i… view at source ↗
read the original abstract

Scanning Photothermal Radiometry (SPR) is an active thermal technique that is simultaneously non-destructive, contactless, and allows for temporal resolutions on the order of nanoseconds, spatial resolutions down to the sub-micrometre scale and at different depths. This scanning method can be time consuming thus this work shows that it is possible to reduce the amount of measurements taken by 6 when using SPR on a sample consisting of carbon fibres in an aluminium matrix. It uses irregular sampling on sparse signals, and a weighted random technique to further decrease the amount of samples needed.

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

1 major / 1 minor

Summary. The manuscript claims that Scanning Photothermal Radiometry (SPR) on a carbon-fibre/aluminium-matrix sample can be performed with a factor-of-6 reduction in the number of measurements by applying irregular sampling to sparse thermal signals together with a weighted random undersampling scheme.

Significance. If the reconstruction fidelity is shown to be preserved, the approach would materially shorten acquisition times for a contactless, high-resolution thermal characterisation technique that already offers nanosecond temporal and sub-micrometre spatial resolution, thereby increasing its utility for non-destructive evaluation of composites.

major comments (1)
  1. [Abstract] Abstract: the central claim of accurate thermal-property recovery at 6 imes undersampling is unsupported by any reported error metric (RMSE, bias in diffusivity or conductivity), sparsity measure (number of significant coefficients), or direct comparison between full and undersampled reconstructions on the same specimen; this quantitative link is load-bearing for the stated reduction factor.
minor comments (1)
  1. [Abstract] The abstract does not name the sparsifying basis or the precise weighting rule used in the random selection; these details should be stated explicitly even at the abstract level.

Simulated Author's Rebuttal

1 responses · 0 unresolved

We thank the referee for their detailed review and constructive comments on our manuscript. We address the major comment below.

read point-by-point responses

  1. Referee: [Abstract] Abstract: the central claim of accurate thermal-property recovery at 6 times undersampling is unsupported by any reported error metric (RMSE, bias in diffusivity or conductivity), sparsity measure (number of significant coefficients), or direct comparison between full and undersampled reconstructions on the same specimen; this quantitative link is load-bearing for the stated reduction factor.

    Authors: We thank the referee for this observation. The manuscript presents the undersampling scheme and resulting thermal maps but does not include the quantitative error metrics or sparsity measures mentioned. We agree that these would strengthen the central claim of accurate recovery at 6x reduction and will add RMSE, bias in diffusivity/conductivity, sparsity measures, and direct quantitative comparisons between full and undersampled cases in the revised manuscript. revision: yes

Circularity Check

0 steps flagged

No circularity: empirical undersampling technique presented without self-referential derivations or fitted predictions

full rationale

The paper describes an experimental reduction in SPR measurements by a factor of 6 via irregular/weighted-random sampling on a carbon-fibre/aluminium specimen. No equations, parameter fits, self-citations, or uniqueness theorems appear in the supplied abstract or description. The central claim is an empirical observation of reconstruction feasibility under a sparsity assumption; it does not reduce any derived quantity to its own inputs by construction. The derivation chain is therefore self-contained against external benchmarks and receives the default non-circularity finding.

Axiom & Free-Parameter Ledger

0 free parameters · 0 axioms · 0 invented entities

Abstract-only review supplies no information on free parameters, axioms or invented entities.

pith-pipeline@v0.9.1-grok · 5666 in / 1001 out tokens · 26042 ms · 2026-06-27T05:52:40.444637+00:00 · methodology

discussion (0)

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

Reference graph

Works this paper leans on

26 extracted references · 15 canonical work pages

  1. [1]

    Sizing the depth and width of ideal delaminations using modulated photothermal radiometry,

    A. Salazar and A. Mendioroz, “Sizing the depth and width of ideal delaminations using modulated photothermal radiometry,” J. Appl. Phys. , vol. 131, no. 8, p. 085 106, 2022-02-23, ISSN : 0021-8979. DOI: 10.1063/5.0085178

    work page doi:10.1063/5.0085178 2022

  2. [2]

    Spatially localized measurement of isotropic and aniso- tropic thermophysical properties by photothermal radiometry,

    G. Hamaoui, E. Villarreal, H. Ban et al., “Spatially localized measurement of isotropic and aniso- tropic thermophysical properties by photothermal radiometry,” Journal of Applied Physics , 2020. DOI: 10.1063/5.0020411. 8 A PREPRINT - 12 TH JUNE 2026

    work page doi:10.1063/5.0020411 2020

  3. [3]

    Measurement of High-temperature Thermophysical Prop- erties of Bulk and Coatings Using Modulated Photothermal Radiometry,

    J. Zeng, K. M. Chung, Q. Wang et al., “Measurement of High-temperature Thermophysical Prop- erties of Bulk and Coatings Using Modulated Photothermal Radiometry,” International Journal of Heat and Mass Transfer , vol. 170, p. 120 989, 2021-05-01, ISSN : 0017-9310. DOI: 10 . 1016 / j . ijheatmasstransfer.2021.120989

    arXiv 2021

  4. [4]

    Thermal characterization of vertical interface by scanning photothermal radiometry,

    A. Mateos-Canseco, A. Kusiak, J.-L. Battaglia et al., “Thermal characterization of vertical interface by scanning photothermal radiometry,” Review of Scientific Instruments, vol. 95, no. 10, p. 104 901, 2024-10-01, ISSN : 0034-6748, 1089-7623. DOI: 10.1063/5.0225690

    work page doi:10.1063/5.0225690 2024

  5. [5]

    Thermal imaging by scanning photothermal radiometry,

    A. Mateos-Canseco, A. Kusiak and J.-L. Battaglia, “Thermal imaging by scanning photothermal radiometry,” Review of Scientific Instruments, vol. 94, no. 10, p. 104 902, 2023-10-01, ISSN : 0034-6748, 1089-7623. DOI: 10.1063/5.0165057

    work page doi:10.1063/5.0165057 2023

  6. [6]

    Photothermal Radiometry,

    P .-E. Nordal and S. O. Kanstad, “Photothermal Radiometry,” Physica Scripta, 1979. DOI: 10.1088/ 0031-8949/20/5-6/020

    1979

  7. [7]

    Near-Optimal Signal Recovery From Random Projections: Universal En- coding Strategies?

    E. J. Candes and T. Tao, “Near-Optimal Signal Recovery From Random Projections: Universal En- coding Strategies?” IEEE Trans. Inform. Theory, vol. 52, no. 12, pp. 5406–5425, 2006-12, ISSN : 0018-

    2006

  8. [8]

    DOI: 10.1109/TIT.2006.885507

    work page doi:10.1109/tit.2006.885507 2006

  9. [9]

    Compressed sensing,

    D. Donoho, “Compressed sensing,” IEEE Trans. Inform. Theory, vol. 52, no. 4, pp. 1289–1306, 2006- 04, ISSN : 0018-9448. DOI: 10.1109/TIT.2006.871582

    work page doi:10.1109/tit.2006.871582 2006

  10. [10]

    Thermal-property microscopy with compressive-sensing frequency- domain thermoreflectance,

    H. Yang, Z. Zhu, Z. Xie et al., “Thermal-property microscopy with compressive-sensing frequency- domain thermoreflectance,”Physical Review Applied, vol. 24, no. 1, p. 014 033, 2025-07-16,ISSN : 2331-

    2025

  11. [11]

    DOI: 10.1103/jqgf-zgtq

    work page doi:10.1103/jqgf-zgtq

  12. [12]

    Implementing an accurate and rapid sparse sampling ap- proach for low-dose atomic resolution STEM imaging,

    L. Kovarik, A. Stevens, A. Liyu et al., “Implementing an accurate and rapid sparse sampling ap- proach for low-dose atomic resolution STEM imaging,” Applied Physics Letters , vol. 109, no. 16, 2016-10-17, ISSN : 0003-6951. DOI: 10.1063/1.4965720

    work page doi:10.1063/1.4965720 2016

  13. [13]

    Fast AFM Imaging Based on Compressive Sensing Using Undersampled Ras- ter Scan,

    Y. Niu and G. Han, “Fast AFM Imaging Based on Compressive Sensing Using Undersampled Ras- ter Scan,” IEEE Transactions on Instrumentation and Measurement , 2021. DOI: 10.1109/TIM.2020. 3023215

    work page doi:10.1109/tim.2020 2021

  14. [14]

    Stable Signal Recovery from Incomplete and Inaccurate Meas- urements

    E. Candes, J. Romberg and T. Tao. “Stable Signal Recovery from Incomplete and Inaccurate Meas- urements.” arXiv: math/0503066. (2005-12-07), pre-published

    Pith/arXiv arXiv 2005

  15. [15]

    Multiquadric equations of topography and other irregular surfaces,

    R. L. Hardy, “Multiquadric equations of topography and other irregular surfaces,” Journal of Geophysical Research , vol. 76, no. 8, pp. 1905–1915, 1971-03-10, ISSN : 2156-2202. DOI: 10 . 1029 / JB076i008p01905

    1905

  16. [16]

    Recent Progress in Modulated Photothermal Radiometry,

    J. Corona and N. Kandadai, “Recent Progress in Modulated Photothermal Radiometry,” Sensors, vol. 23, no. 10, p. 4935, 2023-01, ISSN : 1424-8220. DOI: 10.3390/s23104935

    work page doi:10.3390/s23104935 2023

  17. [17]

    S. L. Brunton and J. N. Kutz, Data-Driven Science and Engineering: Machine Learning, Dynamical Sys- tems, and Control. Cambridge University Press, 2019

    2019

  18. [18]

    Snapshot Compressive Imaging: Theory, Algorithms, and Applications,

    X. Yuan, D. J. Brady and A. K. Katsaggelos, “Snapshot Compressive Imaging: Theory, Algorithms, and Applications,” IEEE Signal Process. Mag. , vol. 38, no. 2, pp. 65–88, 2021-03, ISSN : 1053-5888, 1558-0792. DOI: 10.1109/MSP.2020.3023869

    work page doi:10.1109/msp.2020.3023869 2021

  19. [19]

    Scalable training of L\(

    G. Andrew and J. Gao, “Scalable training of L1-regularized log-linear models,” Proceedings of the 24th international conference on Machine learning, 2007. DOI: 10.1145/1273496.1273501

    work page doi:10.1145/1273496.1273501 2007

  20. [20]

    Compressed Sensing in Python,

    R. Taylor. “Compressed Sensing in Python,” Robert Taylor. (), [Online]. Available: https : / / robert-taylor.me/blog/compressed-sensing-python/ (visited on 2026-05-05)

    2026

  21. [21]

    Undersampled flying spot thermography using com- pressive sensing,

    F. Crouau, J. Maire, J.-L. Battaglia et al. , “Undersampled flying spot thermography using com- pressive sensing,” NDT & E International , vol. 163, p. 103 780, 2026-08-01, ISSN : 0963-8695. DOI: 10.1016/j.ndteint.2026.103780

    work page doi:10.1016/j.ndteint.2026.103780 2026

  22. [22]

    Spatial interpolation: An analytical comparison between kriging and RBF networks,

    V . S. Fazio and M. Roisenberg, “Spatial interpolation: An analytical comparison between kriging and RBF networks,” in Proceedings of the 28th Annual ACM Symposium on Applied Computing, Coim- bra Portugal: ACM, 2013-03-18, pp. 2–7, ISBN : 978-1-4503-1656-9. DOI: 10.1145/2480362.2480364

    work page doi:10.1145/2480362.2480364 2013

  23. [23]

    Interpolation of scattered data: Distance matrices and conditionally positive def- inite functions,

    C. A. Micchelli, “Interpolation of scattered data: Distance matrices and conditionally positive def- inite functions,” Constr. Approx, vol. 2, no. 1, pp. 11–22, 1986-12-01, ISSN : 1432-0940. DOI: 10.1007/ BF01893414

    1986

  24. [24]

    G. E. Fasshauer, Meshfree Approximation Methods with MATLAB (Interdisciplinary Mathematical Sciences v. 6). Singapore ; Hackensack, N.J: World Scientific, 2007, 500 pp.,ISBN : 978-981-270-634-8 978-981-270-633-1. 9 A PREPRINT - 12 TH JUNE 2026

    2007

  25. [25]

    RBFInterpolator — SciPy v1.17.0 Manual

    “RBFInterpolator — SciPy v1.17.0 Manual.” (), [Online]. Available: https://docs.scipy.org/ doc/scipy/reference/generated/scipy.interpolate.RBFInterpolator.html (visited on 2026- 05-06)

    2026

  26. [26]

    Chlan, K

    U. Seidel, K. Haupt, H. G. Walther et al., “An attempt towards quantitative photothermal micro- scopy,” J. Appl. Phys., vol. 78, no. 3, pp. 2050–2056, 1995-08-01, ISSN : 0021-8979. DOI: 10.1063/1. 360182. 10

    work page doi:10.1063/1 2050