Pith. sign in

REVIEW 3 major objections 5 minor 56 references

Deciphering hidden layers images through terahertz spectral fingerprints

T0 review · 3 major / 5 minor · reviewed 2026-08-07 · deepseek-v4-flash

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗

Pith's one-line read Terahertz multispectral imaging deciphers hidden images beneath layered paint by matching each pixel's spectrum to pigment-specific absorption fingerprints.

desk verdict A plausible THz spectral-mapping proof-of-concept whose main claims are weakened by manual post-hoc pixel correction and an untested assumption that pellet transmission fingerprints survive in reflection from painted layers. read the letter →

arxiv 2505.23965 v1 pith:LVD6CBFA submitted 2025-05-29 physics.optics physics.app-phphysics.ins-det

Candida Moffa , Daniele Francescone , Alessandro Curcio , Anna Candida Felici , Marco Bellaveglia , Luca Piersanti , Mauro Migliorati , Massimo Petrarca This is my paper
classification physics.opticsphysics.app-phphysics.ins-det
keywords terahertzspectroscopymultispectralimagingpigmentidentificationculturalheritagehiddenlayersnon-destructiveanalysissparsedeconvolutionstratigraphy
verification ladder T0 review T1 audit T2 compute T3 formal

The pith

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

The reading

The paper reports a terahertz (THz) reflection-imaging method that reconstructs images hidden under paint layers and identifies the pigment in each layer without damaging the object. The authors first measure transmission spectra of three historical pigments—cinnabar, orpiment, and realgar—then raster-scan painted paper mock-ups in reflection, Fourier-transforming each pixel and integrating the signal around each pigment's characteristic absorption peak. On a three-layer mock-up, the method recovers the bottom letter through a fully covering layer, distinguishes two chemically similar arsenic-sulfide pigments, and estimates each layer's thickness; results are checked against optical microscopy and sparse-deconvolution analysis. The significance is a single-measurement, chemical, depth-resolved readout for layered artworks, something X-ray or Raman approaches do only under narrower conditions.

What carries the argument

The central object is the spectral fingerprint: a pigment-specific absorption peak in the 0.5–3 THz range, such as cinnabar near 1.13 THz, orpiment near 1.59 THz, and realgar near 1.48 THz. The accompanying machinery is frequency-domain chemical mapping: each time-domain pixel waveform is Fourier-transformed, the spectrum is integrated over a narrow window around the fingerprint, and the integrated signal is compared with the same integral on bare paper via the threshold $L_{\text{signal}} \geq 3\,L_{\text{reference}}$. Sparse deconvolution with $\ell^1$ regularization then turns the reflected pulse train into distinct interface echoes, and thickness follows from echo spacing divided by the material's refractive index.

What would settle it

Scan a mock-up whose hidden composition is known, run the algorithm blind, and compare the reconstructed maps pixel-by-pixel with the exposed layers; if a covered pigment no longer satisfies $L_{\text{signal}} \geq 3\,L_{\text{reference}}$ at its pellet-measured peak frequency, the central claim fails. A direct test is to measure a cinnabar patch before and after covering it with orpiment and check whether the 1.13 THz integrated area remains above the bare-paper threshold.

Watch

Extended reading notes

Core claim

The central claim is that THz multispectral imaging, using spectral fingerprints measured in transmission on pure pigments, can be applied in reflection to layered painted samples to produce per-pixel chemical maps: a pixel is assigned to a pigment when the area under its reflected spectrum over the pigment's absorption band is at least three times the area measured on bare paper. On the three mock-ups, this custom algorithm separates letter from background, separates orpiment from realgar despite their similar elemental composition, and, for the three-layer sample, recovers each of the three superimposed letters from a single raster scan. The same scan feeds a sparse-deconvolution routine that locates layer interfaces and, with the measured refractive indices, gives thicknesses of the paint layers that agree with digital-microscope cross-sections.

Load-bearing premise

The method assumes that the terahertz absorption pattern of each pigment, measured on a pressed pellet of pure powder in transmission, stays the same when the pigment is painted as a thin layer with gum Arabic on paper and viewed in reflection, even through other paint layers.

Editorial extensions

If this is right

  • A single THz reflection scan of a layered painting can yield, at once, the hidden image at each depth, the pigment identity of each layer, and an estimate of each layer's thickness.
  • Pigments that X-ray fluorescence cannot separate because they share elements—here orpiment and realgar, both arsenic sulfides—can be told apart by their THz fingerprints.
  • Because binders and varnishes show no strong THz fingerprints, pigment assignment is not confused by the medium the pigment is painted in.
  • Frequency-domain chemical maps give markedly better contrast for concealed layers than maps built from the time-domain waveform's maximum, minimum, or peak-to-peak amplitude.
  • Layer thicknesses retrieved from sparse-deconvolution echo times agree with optical-microscope cross-sections within the stated uncertainties.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

  • Beyond the paper's results, the 3× reference threshold and the manual pixel corrections could likely be replaced by a per-pixel classifier trained on the same band-integral features, which might reduce the reported ~13% discrepancy without visual inspection.
  • An extension not explored in the paper is to build a reflection-geometry spectral library of pigments in realistic binders, measuring peak shifts relative to pressed-pellet transmission data; that would turn the mock-up demonstration into a field-ready protocol.
  • Because the measured beam waist limits spatial resolution to about 1 mm, the same frequency-domain contrast mechanism should, with tighter focusing or post-processing super-resolution, be able to read smaller hidden script.
  • The method's logic applies to any stratified object with a stable THz absorption band, so layered polymers, coatings, or biological samples are plausible testbeds for the same algorithm.
Share X Bluesky LinkedIn Reddit HN

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Referee Report

3 major / 5 minor

Summary. The manuscript presents a THz time-domain spectroscopy (THz-TDS) study of three painted paper mock-ups. The authors build a transmission-mode spectral database of three arsenic/sulfur pigments (cinnabar, orpiment, realgar), then raster-scan the samples in reflection geometry, Fourier-transform the time-domain waveforms, integrate the reflected spectral intensity over narrow bands around the database peaks, and apply a threshold condition (L_signal ≥ 3·L_reference, Eq. 3) to produce 2D pigment maps. For the three-layer sample S3, they additionally use sparse deconvolution to estimate layer thicknesses, which they compare with optical microscopy cross-sections. The central claims are that hidden images can be reconstructed, layer thicknesses estimated, and pigments identified simultaneously from a single THz scan.

Significance. If the central claims hold, the method would be a useful non-invasive tool for cultural-heritage diagnostics, with the notable strength of distinguishing chemically similar arsenic sulfides (orpiment vs. realgar) that are difficult for XRF. The work benefits from an explicit comparison of sparse-deconvolution thickness estimates with digital-microscopy cross-sections, and the pigment database extends previously reported fingerprints. However, the load-bearing demonstration is weakened by two issues: the unvalidated transfer of transmission-pellet fingerprints to reflection measurements of painted/buried layers, and the post-hoc manual correction of the algorithm output using knowledge of the true letter shapes. As presented, the quantitative discrepancy figures are in-sample, and the final maps are not an independent test of the algorithm. These issues are fixable but require additional experiments and a more disciplined evaluation protocol.

major comments (3)
  1. [Sections 2.1, 2.2, Eq. (3)] The classification pipeline assumes that absorption fingerprints measured in transmission on pressed pure-powder pellets (Section 2.1) remain observable, unshifted, and separable in reflection spectra of thin gum-Arabic-bound paint layers on paper, including layers buried beneath other pigments. No spectral comparison is shown between the pellet transmission spectra and the reflection spectra of painted layers, nor is there a test of the buried orpiment layer under the cinnabar cover in S3. In reflection, I_ij(ν) contains Fresnel reflection and Fabry–Perot contributions, and the reference used in Eq. (3) is bare paper, not the multilayer stack that overlies the buried layer. The authors should validate the fingerprint transferability by acquiring reflection spectra of single painted layers and of the actual S3 stack (or a controlled stack replica), demonstrating that the database peak positions remain identifiable and that the threshold in Eq. (3) is physically meaningful. Without this, the central material-identification claim is not supported.
  2. [Section 2.2 (paragraph on visual inspection) and Figures 3, 4, 6] The reported maps are post-correction outputs. The text states that each two-dimensional map underwent visual inspection to correct 'overestimated' and 'underestimated' pixels, and the corrections are guided by the known letter shapes (T, H, Z, C) from sample preparation. Therefore, the discrepancy percentages (e.g., 12.7% overestimation for realgar in S3) describe the raw algorithm output, not the final reconstructed images, and the final images are not independent evidence of deciphering concealed text. The authors should report precision/recall or pixel-wise error for the raw algorithm output against ground-truth masks, and state separately how many pixels were manually changed for each map. If manual correction is retained, its criteria must be documented in a way that does not rely on the expected letter shape.
  3. [Section 4.2, Eq. (1) and Eq. (3)] The center frequency ν0 in Eq. (1) is chosen by maximizing contrast on the same dataset used to produce the maps, and the threshold factor 3 in Eq. (3) is introduced without a sensitivity analysis. Consequently, the reported discrepancy values are in-sample and do not demonstrate predictive validity. The authors should provide a sensitivity analysis over ν0, the integration half-width Δν, and the threshold factor, and ideally evaluate the algorithm on an independent subset of pixels (e.g., a train/test split of the raster scan or a separate mock-up). The choice Δν = 0.02 THz is also not justified in Eq. (1), which only defines ν0; the role of Δν in the integration of Eq. (2) should be clarified.
minor comments (5)
  1. [Section 1] There is a typo in the second paragraph: 'while everal methodologies' should read 'while several methodologies'.
  2. [Section 2.2 and Figure 2] The references to Figure 2 panels in the text are inconsistent with the caption: the text says the minimum-amplitude map for S1 is Figure 2b and the maximum-amplitude map for S2 is Figure 2d, but the caption identifies Figure 2b as S2 (maximum) and Figure 2d as S1 (minimum). Please correct the cross-references.
  3. [Section 2.2.1] In the sentence describing the cinnabar map of S1, the text says the corrected pixels were verified by 'visually checking each spectra (Fig. 3a)', but Figure 3a is the visible image of the sample; the reference likely should be to the THz chemical map in Figure 3b or 3c.
  4. [Section 2.2 and Section 3] There are minor language errors such as 'compostion' (Section 2.2) and 'image image' (Section 3, Discussion); please proofread for these and similar issues.
  5. [Section 4.4] The phrase 'pressed into containment bolts' is unclear; presumably the powder was pressed into pellet holders or sample cells. Please use standard terminology.

Circularity Check

2 steps flagged · score 6.0 of 10

Hidden-layer reconstructions are partly constructed from ground truth: the detection band is optimized on the same data and the maps are manually corrected against known letter layouts.

  1. fitted input called prediction [Section 2.2 (Mock-ups mapping) and Section 4.2, Eq. (1)]
    "Then, a frequency range that includes the absorption peak of interest ( ν0 ± ∆ν) is chosen to maximize the image contrast as: max||(Iref erence(ν0) − Isignal(v0)|| (1)"

    The spectral integration band used to classify each pixel is not fixed independently by the pigment database; it is selected by maximizing the contrast between reference and sample spectra on the same pixels that are subsequently classified. Thus the contrast in the final chemical maps is optimized on the evaluation data themselves. The reported image quality and the pixel-discrepancy percentages are therefore in-sample fitting results rather than independent out-of-sample predictions of pigment presence.

  2. other [Section 2.2 (Mock-ups mapping), correction paragraph]
    "Each two-dimensional map underwent visual inspection to identify pixels exhibiting potential algorithmic output errors (i.e., "false assessment"). This corrective procedure aimed to mitigate "overestimated" pixels, wherein the algorithm falsely identified pigment presence, and to rectify "underestimated" pixels, where the algorithm failed to detect the pigment’s characteristic peak."

    The expected letter shapes in S1, S2, and S3 were known from sample preparation (Table 2 and Fig. 9), including the hidden orpiment 'Z' in S3. Calling a pixel 'underestimated' or 'overestimated' therefore required comparing the algorithm output with the known ground-truth layout. Underestimated pixels were replaced by nearest-neighbor means and overestimated pixels were zeroed, so the final maps in Figs. 3, 4, and 6 incorporate the expected images by construction. The claimed reconstruction of hidden layers and the ~13% discrepancy statistic are thus not purely derived from THz spectra; they include manual, ground-truth-guided editing.

full rationale

The main physical assumption—that transmission pellet fingerprints survive in reflection geometry on thin painted layers with gum Arabic binder—is not validated in the paper, but that is an external-validity/correctness risk rather than circularity, and it is not scored as circular here. The circularity concerns the evaluation of the central claim. The detection band ν0 is chosen by Eq. (1) to maximize contrast on the same data that are then mapped, and the reconstructed maps are manually corrected against known letter layouts, including the hidden base layer. Therefore the reported reconstruction and the ~13% discrepancy are in-sample, partially hand-edited outcomes rather than independent predictions. The pigment database itself is supported by external literature ([11,13,28,42,43]) and the sparse-deconvolution thicknesses are checked against optical microscopy, so the paper is not wholly circular. Because the central 'prediction' is partly constructed from its own inputs, a moderate score of 6 is appropriate.

Assumptions & free parameters 3 free parameters · 3 assumptions · 0 invented entities

The central mapping depends on a threshold factor and integration bands that are chosen ad hoc and fit to the target images, plus a sparse deconvolution parameter that is not reported. The assumptions about spectral transferability and constant refractive indices are standard in the field but not independently validated here.

free parameters (3)
  • Threshold factor = 3
    The factor multiplying L_reference in Eq. 3 is chosen to reduce false assessments; no optimization or justification beyond 'three times' is given.
  • Integration band widths = ν0 ± 0.02 THz (e.g., cinnabar at 1.13 THz)
    The band is selected for each pigment to maximize contrast (Eq. 1); the width is reported as ±0.02 THz in the examples, but the choice is data-dependent.
  • Sparse deconvolution regularization λ = not reported
    The L1 regularization parameter in Eq. 6 controls the sparsity and echo detection; without its value the thickness retrieval cannot be reproduced.
assumptions (3)
  • domain assumption THz absorption peaks of pigments measured in pressed pellets are preserved when the pigments are painted with a binder on paper and measured in reflection through overlying layers.
    This transfers the spectral database from Section 2.1 to the mapping in Section 2.2 without a quantitative validation.
  • domain assumption The reflected THz waveform is a linear convolution of the incident pulse with the system response (Eq. 4-5), and the incident pulse can be measured from a metal mirror.
    Standard model for THz-TDS, used in the sparse deconvolution.
  • domain assumption The refractive indices of the pigments and paper are constant across the THz band and equal to the values measured on pure pellets.
    Used to convert echo delays to thickness in Section 2.2.2.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Deciphering hidden layers images through terahertz spectral fingerprints." pith.science (2026). https://pith.science/paper/LVD6CBFA

@misc{pith2026250523965,
  author       = {Pith},
  title        = {Pith review of: Deciphering hidden layers images through terahertz spectral fingerprints},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/LVD6CBFA}},
  note         = {Machine review of arXiv:2505.23965}
}
read the original abstract

Terahertz radiation enables non destructive, depthresolved analysis of layered artworks. This study demonstrates THz multispectral imaging ability to reveal concealed text beneath mockup of pictorial layers, reconstructing hidden narratives at varying depths through frequency domain analysis.Simultaneously, it maps pigment composition, providing valuable chemical information. Showcasing the power to penetrate and decipher stratified materials, this work establishes THz multispectral imaging as a crucial tool for unlocking hidden secrets and characterizing materials in Cultural Heritage artifacts.

Figures

Figures reproduced from arXiv: 2505.23965 by the authors.

Figure 1
Figure 1. Lorentzian fitting of the absorbance spectrum of pure pigments a) cinnabar, b) [PITH_FULL_IMAGE:figures/full_fig_p006_1.png] view at source ↗
Figure 2
Figure 2. Image obtained from the maximum of the THz temporal waveforms for a) S1, [PITH_FULL_IMAGE:figures/full_fig_p007_2.png] view at source ↗
Figure 3
Figure 3. a) Visible image of sample S1 letter ”T” in cinnabar and background of realgar; [PITH_FULL_IMAGE:figures/full_fig_p010_3.png] view at source ↗
Figures from the paper (6 more)
Figure 4
Figure 4. Figure 4: a) Visible image of sample S2 letter ”H” in realgar and background of orpiment; [PITH_FULL_IMAGE:figures/full_fig_p011_4.png]
Figure 5
Figure 5. Figure 5: Raw signal and sparse deconvolved signals for one of the highly absorbing points [PITH_FULL_IMAGE:figures/full_fig_p012_5.png]
Figure 6
Figure 6. Figure 6: Sample constituted by superimposed layers (S3): a) visible image of base layer [PITH_FULL_IMAGE:figures/full_fig_p014_6.png]
Figure 7
Figure 7. Figure 7: Microphotograph of a cross-sectional layering of sample S3 in which are reported [PITH_FULL_IMAGE:figures/full_fig_p015_7.png]
Figure 8
Figure 8. Figure 8: Schematic representation of the proposed set-up for the investigation of paper [PITH_FULL_IMAGE:figures/full_fig_p018_8.png]
Figure 9
Figure 9. Figure 9: Schematic structure of the multilayered structure of samples S3 representing [PITH_FULL_IMAGE:figures/full_fig_p021_9.png]

Discussion (0). Sign in to comment.

Reference graph

Works this paper leans on

56 extracted references · 56 canonical work pages

  1. [1]

    Frosch, K

    T. Frosch, K. A. Chan, H. C. Wong, J. T. Cabral, S. G. Kazarian, Nondestructive three-dimensional analysis of layered polymer structures with chemical imaging, Langmuir 26 (24) (2010) 19027–19032

    work page 2010

  2. [2]

    V. P. Zharov, S. Ferguson, J. F. Eidt, P. C. Howard, L. M. Fink, M. Waner, Infrared imaging of subcutaneous veins, Lasers in Surgery and Medicine: The Official Journal of the American Society for Laser Medicine and Surgery 34 (1) (2004) 56–61

    work page 2004

  3. [3]

    Conti, M

    C. Conti, M. Realini, C. Colombo, K. Sowoidnich, N. K. Afseth, M. Bertasa, A. Botteon, P. Matousek, Noninvasive analysis of thin tur- bid layers using microscale spatially offset raman spectroscopy, Analyt- ical chemistry 87 (11) (2015) 5810–5815

    work page 2015

  4. [4]

    Tonazzini, E

    A. Tonazzini, E. Salerno, Z. A. Abdel-Salam, M. A. Harith, L. Marras, A. Botto, B. Campanella, S. Legnaioli, S. Pagnotta, F. Poggialini, et al., Analytical and mathematical methods for revealing hidden details in an- cient manuscripts and paintings: A review, Journal of advanced research 17 (2019) 31–42

    work page 2019

  5. [5]

    Perino, L

    M. Perino, L. Pronti, C. Moffa, M. Rosellini, A. C. Felici, New frontiers in the digital restoration of hidden texts in manuscripts: A review of the technical approaches, Heritage 7 (2) (2024)

    work page 2024

  6. [6]

    Conti, C

    C. Conti, C. Colombo, M. Realini, P. Matousek, Subsurface analysis of painted sculptures and plasters using micrometre-scale spatially offset raman spectroscopy (micro-sors), Journal of Raman Spectroscopy 46 (5) (2015) 476–482

    work page 2015

  7. [7]

    Botteon, C

    A. Botteon, C. Conti, M. Realini, C. Colombo, P. Matousek, Discovering hidden painted images: subsurface imaging using microscale spatially offset raman spectroscopy, Analytical chemistry 89 (1) (2017) 792–798. 20

    work page 2017

  8. [8]

    Vermeulen, C

    M. Vermeulen, C. Conti, P. Matousek, L. V. Angelova, A. Botteon, Un- ravelling hidden text and figures in paper-based archival documents with micro-spatially offset raman spectroscopy imaging, Spectrochimica Acta Part A: Molecular and Biomolecular Spectroscopy 329 (2025) 125591

    work page 2025

Show all 56 references
  1. [9]

    Conti, C

    C. Conti, C. Colombo, M. Realini, G. Zerbi, P. Matousek, Subsurface raman analysis of thin painted layers, Applied spectroscopy 68 (6) (2014) 686–691

    2014

  2. [10]

    Conti, A

    C. Conti, A. Botteon, C. Colombo, M. Realini, P. Matousek, Investiga- tion of heterogeneous painted systems by micro-spatially offset raman spectroscopy, Analytical chemistry 89 (21) (2017) 11476–11483

    2017

  3. [11]

    Y. Yang, D. Zhai, Z. Zhang, C. Zhang, Thz spectroscopic identification of red mineral pigments in ancient chinese artworks, Journal of Infrared, Millimeter, and Terahertz Waves 38 (2017) 1232–1240

    2017

  4. [12]

    E. M. Kleist, C. L. Koch Dandolo, J.-P. Guillet, P. Mounaix, T. M. Korter, Terahertz spectroscopy and quantum mechanical simulations of crystalline copper-containing historical pigments, The Journal of Phys- ical Chemistry A 123 (6) (2019) 1225–1232

    2019

  5. [13]

    E. M. Kleist, T. M. Korter, Quantitative analysis of minium and vermil- ion mixtures using low-frequency vibrational spectroscopy, Analytical chemistry 92 (1) (2019) 1211–1218

    2019

  6. [14]

    Moffa, C

    C. Moffa, C. Merola, E. Chiadroni, L. Giuliano, A. Curcio, L. Palumbo, A. C. Felici, M. Petrarca, Pigments, minerals, and copper-corrosion products: Terahertz continuous wave (thz-cw) spectroscopic characteri- zation of antlerite and atacamite, Journal of Cultural Heritage 66 ...

    2024

  7. [15]

    Moffa, A

    C. Moffa, A. Curcio, C. Merola, M. Migliorati, L. Palumbo, A. C. Felici, M. Petrarca, Discrimination of natural and synthetic forms of azurite: an innovative approach based on high-resolution terahertz continuous wave (thz-cw) spectroscopy for cultural heritage., Dyes and Pigm...

    2024

  8. [16]

    Moffa, V

    C. Moffa, V. Urso, M. Migliorati, L. Palumbo, A. C. Felici, G. Zollo, M. Petrarca, Terahertz time-domain investigation of atacamite: Spectral 21 analysis and theoretical insights for cultural heritage applications, Ap- plied Physics Letters 125 (20) (2024) 204102. doi:10.1063/...

    2024 doi

  9. [17]

    Moffa, A

    C. Moffa, A. C. Felici, M. Petrarca, Terahertz investigation of cultural heritage synthetic materials: A case study of copper silicate pigments, Minerals 15 (5) (2025). doi:10.3390/min15050490

    2025 doi

  10. [18]

    Fukunaga, I

    K. Fukunaga, I. Hosako, I. Duling III, M. Picollo, Terahertz imaging systems: a non-invasive technique for the analysis of paintings, in: O3A: Optics for Arts, Architecture, and Archaeology II, Vol. 7391, SPIE, 2009, pp. 119–127

    2009

  11. [19]

    D. M. Mittleman, S. Hunsche, L. Boivin, M. C. Nuss, T-ray tomography, Optics letters 22 (12) (1997) 904–906

    1997

  12. [20]

    A. J. Fitzgerald, B. E. Cole, P. F. Taday, Nondestructive analysis of tablet coating thicknesses using terahertz pulsed imaging, Journal of pharmaceutical sciences 94 (1) (2005) 177–183

    2005

  13. [21]

    C. L. K. Dandolo, K. Fukunaga, Y. Kohzuma, K. Kiriyama, K. Mat- suda, P. U. Jepsen, Inspection of asian lacquer substructures by ter- ahertz time-domain imaging (thz-tdi), Journal of Infrared, Millimeter, and Terahertz Waves 38 (2017) 425–434

    2017

  14. [22]

    St¨ ubling, A

    E.-M. St¨ ubling, A. Rehn, T. Siebrecht, Y. Bauckhage, L. ¨Ohrstr¨ om, P. Eppenberger, J. C. Balzer, F. R¨ uhli, M. Koch, Application of a robotic thz imaging system for sub-surface analysis of ancient human remains, Scientific reports 9 (1) (2019) 3390

    2019

  15. [23]

    Duling, D

    I. Duling, D. Zimdars, Revealing hidden defects, Nature Photonics 3 (11) (2009) 630–632

    2009

  16. [24]

    Fukunaga, I

    K. Fukunaga, I. Hosako, Y. Kohdzuma, T. Koezuka, M.-J. Kim, T. Ikari, X. Du, Terahertz analysis of an east asian historical mural painting, Journal of the European Optical Society-Rapid publications 5 (2010)

    2010

  17. [25]

    C. L. Dandolo, P. U. Jepsen, M. C. Christensen, Characterization of european lacquers by terahertz (thz) reflectometric imaging, in: 2013 Digital Heritage International Congress (DigitalHeritage), Vol. 1, IEEE, 2013, pp. 89–94. 22

    2013

  18. [26]

    A. S. Skryl, J. B. Jackson, M. I. Bakunov, M. Menu, G. A. Mourou, Terahertz time-domain imaging of hidden defects in wooden artworks: application to a russian icon painting, Applied optics 53 (6) (2014) 1033– 1038

    2014

  19. [27]

    C. L. Koch-Dandolo, T. Filtenborg, K. Fukunaga, J. Skou-Hansen, P. U. Jepsen, Reflection terahertz time-domain imaging for analysis of an 18th century neoclassical easel painting, Applied optics 54 (16) (2015) 5123– 5129

    2015

  20. [28]

    Picollo, K

    M. Picollo, K. Fukunaga, J. Labaune, Obtaining noninvasive strati- graphic details of panel paintings using terahertz time domain spec- troscopy imaging system, Journal of Cultural Heritage 16 (1) (2015) 73–80

    2015

  21. [29]

    Guillet, M

    J.-P. Guillet, M. Roux, K. Wang, X. Ma, F. Fauquet, H. Balacey, B. Re- cur, F. Darracq, P. Mounaix, Art painting diagnostic before restoration with terahertz and millimeter waves, Journal of Infrared, Millimeter, and Terahertz Waves 38 (2017) 369–379

    2017

  22. [30]

    Catapano, F

    I. Catapano, F. Soldovieri, A data processing chain for terahertz imaging and its use in artwork diagnostics, Journal of Infrared, Millimeter, and Terahertz Waves 38 (2017) 518–530

    2017

  23. [31]

    Cacciari, D

    I. Cacciari, D. Ciofini, L. Gallo, S. Siano, Thz imaging tool for char- acterizing both visible and concealed metals in paint layers, in: IOP Conference Series: Materials Science and Engineering, Vol. 364, IOP Publishing, 2018, p. 012096

    2018

  24. [32]

    Catapano, F

    I. Catapano, F. Soldovieri, Thz imaging and data processing: State of the art and perspective, Innovation in near-surface geophysics (2019) 399–417

    2019

  25. [33]

    Fukunaga, Nondestructive evaluation of lined paintings by thz pulsed time-domain imaging, Heritage 6 (4) (2023) 3448–3460

    K. Fukunaga, Nondestructive evaluation of lined paintings by thz pulsed time-domain imaging, Heritage 6 (4) (2023) 3448–3460

    2023

  26. [34]

    A. J. Adam, P. C. Planken, S. Meloni, J. Dik, Terahertz imaging of hidden paint layers on canvas, Optics Express 17 (5) (2009) 3407–3416

    2009

  27. [35]

    J. B. Jackson, J. Bowen, G. Walker, J. Labaune, G. Mourou, M. Menu, K. Fukunaga, A survey of terahertz applications in cultural heritage 23 conservation science, IEEE Transactions on Terahertz Science and Tech- nology 1 (1) (2011) 220–231

    2011

  28. [36]

    Seco-Martorell, V

    C. Seco-Martorell, V. L´ opez-Dom ´ ınguez, G. Arauz-Garofalo, A. Redo- Sanchez, J. Palacios, J. Tejada, Goya’s artwork imaging with terahertz waves, Optics express 21 (15) (2013) 17800–17805

    2013

  29. [37]

    A. M. Gomez-Sepulveda, A. Hernandez-Serrano, R. Radpour, C. Koch- Dandolo, S. Rojas-Landeros, L. Ascencio-Rojas, A. Zarate, G. Hernan- dez, R. Gonzalez-Tirado, M. Insaurralde-Caballero, et al., History of mexican easel paintings from an altarpiece revealed by non-invasive ter-...

    2017

  30. [38]

    J. Lee, Y. Song, H. Lee, J. Zeitler, Extraction of spectral information from terahertz reflectometry data, in: 2022 47th International Con- ference on Infrared, Millimeter and Terahertz Waves (IRMMW-THz), IEEE, 2022, pp. 1–2

    2022

  31. [39]

    Y.-C. Shen, P. F. Taday, Development and application of terahertz pulsed imaging for nondestructive inspection of pharmaceutical tablet, IEEE Journal of Selected Topics in Quantum Electronics 14 (2) (2008) 407–415

    2008

  32. [40]

    D. M. Charron, K. Ajito, J.-Y. Kim, Y. Ueno, Chemical mapping of pharmaceutical cocrystals using terahertz spectroscopic imaging, Ana- lytical Chemistry 85 (4) (2013) 1980–1984

    2013

  33. [41]

    Bandyopadhyay, A

    A. Bandyopadhyay, A. Sengupta, A review of the concept, applica- tions and implementation issues of terahertz spectral imaging technique, IETE Technical Review 39 (2) (2022) 471–489

    2022

  34. [42]

    Artesani, R

    A. Artesani, R. Lamuraglia, F. Menegazzo, S. Bonetti, A. Traviglia, Terahertz time-domain spectroscopy in reflection configuration for inor- ganic and mineral pigment identification, Applied Spectroscopy 77 (1) (2023) 74–87

    2023

  35. [43]

    Zhang, G.-H

    Y.-W. Zhang, G.-H. Ren, X.-Q. Su, T.-H. Meng, G.-Z. Zhao, Terahertz spectroscopy and lattice vibrational analysis of pararealgar and orpi- ment, Chinese Physics B 31 (10) (2022) 103302. 24

    2022

  36. [44]

    E. S. Reyes-Reyes, R. Carriles-Jaimes, E. D’Angelo, S. Nazir, C. L. Koch-Dandolo, F. Kuester, P. U. Jepsen, E. Castro-Camus, Terahertz time-domain imaging for the examination of gilded wooden artifacts, Scientific Reports 14 (1) (2024) 6261

    2024

  37. [45]

    M. Zhai, D. Citrin, A. Locquet, Terahertz nondestructive stratigraphic analysis of complex layered structures: reconstruction techniques, Jour- nal of Infrared, Millimeter, and Terahertz Waves 42 (2021) 929–946

    2021

  38. [46]

    T. Lei, B. Tobin, Z. Liu, S.-Y. Yang, D.-W. Sun, A terahertz time- domain super-resolution imaging method using a local-pixel graph neu- ral network for biological products, Analytica Chimica Acta 1181 (2021) 338898

    2021

  39. [47]

    F. Yuan, S. Zhou, T. Cheng, Y. Liu, A low-and high-resolution terahertz image pair construction method with gradient fusion for learning-based super-resolution, IEEE Access 10 (2022) 132506–132514

    2022

  40. [48]

    D. A. Vazquez, L. Pilozzi, E. DelRe, C. Conti, M. Missori, Terahertz imaging super-resolution for documental heritage diagnostics, IEEE Transactions on Terahertz Science and Technology (2024)

    2024

  41. [49]

    G. Wang, F. Qi, Super-resolution terahertz imaging algorithm based on blind deconvolution, Applied Optics 64 (10) (2025) 2383–2389

    2025

  42. [50]

    Dal Fovo, S

    A. Dal Fovo, S. Mattana, A. Chaban, D. Quintero Balbas, J. L. Lagarto, J. Striova, R. Cicchi, R. Fontana, Fluorescence lifetime phasor analysis and raman spectroscopy of pigmented organic binders and coatings used in artworks, Applied Sciences 12 (1) (2021) 179

    2021

  43. [51]

    Vandenabeele, H

    P. Vandenabeele, H. G. Edwards, L. Moens, A decade of raman spec- troscopy in art and archaeology, Chemical reviews 107 (3) (2007) 675– 686

    2007

  44. [52]

    Pause, I

    R. Pause, I. D. van der Werf, K. J. van den Berg, Identification of pre- 1950 synthetic organic pigments in artists’ paints. a non-invasive ap- proach using handheld raman spectroscopy, Heritage 4 (3) (2021) 1348– 1365

    2021

  45. [53]

    M. Koch, D. M. Mittleman, J. Ornik, E. Castro-Camus, Terahertz time- domain spectroscopy, Nature Reviews Methods Primers 3 (1) (2023) 48. 25

    2023

  46. [54]

    Prati, E

    S. Prati, E. Joseph, G. Sciutto, R. Mazzeo, New advances in the ap- plication of ftir microscopy and spectroscopy for the characterization of artistic materials, Accounts of chemical research 43 (6) (2010) 792–801

    2010

  47. [55]

    F. Rosi, A. Federici, B. G. Brunetti, A. Sgamellotti, S. Clementi, C. Mil- iani, Multivariate chemical mapping of pigments and binders in easel painting cross-sections by micro ir reflection spectroscopy, Analytical and Bioanalytical Chemistry 399 (2011) 3133–3145

    2011

  48. [56]

    J. Dong, X. Wu, A. Locquet, D. S. Citrin, Terahertz superresolution stratigraphic characterization of multilayered structures using sparse deconvolution, IEEE Transactions on Terahertz Science and Technol- ogy 7 (3) (2017) 260–267. arXiv:hal-02148308, doi:10.1109/TTHZ. 2017.26...

    2017

Pith tools

Reviewed August 7, 2026 · model on record in the stance chip above.