Pith. sign in

REVIEW 3 major objections 7 minor 47 references

Automatic Spectral Calibration of Hyperspectral Images:Method, Dataset and Benchmark

T0 review · 3 major / 7 minor · reviewed 2026-08-11 · deepseek-v4-flash

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

Pith's one-line read This paper argues that a learned transformer can remove global illumination from a single hyperspectral image without a physical reference, predicting calibrated reflectance directly from the raw intensity image.

desk verdict Genuinely new HSI calibration task and dataset, with a modest real-data win and an expansion benchmark whose large margins are confounded by shared illuminants. read the letter →

arxiv 2412.14925 v2 pith:5KTOLOK6 submitted 2024-12-19 cs.CV eess.IV

Zhuoran Du , Shaodi You , Cheng Cheng , Shikui Wei This is my paper · ORCID
classification cs.CVeess.IV
keywords hyperspectralimagecalibrationspectralilluminationtransformerattentionreflectanceestimationwhitebalancingdatasetbenchmark
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

Hyperspectral images need per-band calibration to turn raw digital counts into reflectance, but the standard methods require a white panel in the scene or two successive captures with a fixed camera. This paper argues that a learned map can do the job automatically: given a single uncalibrated image, the Spectral Illumination Transformer (SIT) predicts the calibrated reflectance directly. To support that claim, the authors built the first dataset for this task, 765 real uncalibrated/calibrated pairs across natural scenes, expanded to 7,650 pairs by replaying ten physically measured illuminations. On their benchmark, SIT outperforms Gray-World and three learning-based restoration baselines on PSNR, SAM, RMSE, and ERGAS, with the illumination-attention module contributing most of the gain. The benchmark also shows low-light scenes and near-infrared bands remain the hardest cases, which the paper leaves to future work.

What carries the argument

The load-bearing component is the Spectral Illumination Transformer unit (SIT-U), which runs two parallel attention branches inside each layer of a U-shaped encoder-decoder. The spectral-attention branch is the standard HSI transformer branch; the illumination-attention branch downsamples the feature map with strided convolutions and average pooling, averages over all spatial positions to extract a per-channel global illumination feature, projects that feature into query and key vectors, and forms an illumination attention matrix $A_I$. The final attention weights are $A_{SI}=\operatorname{softmax}(A_S \cdot A_I)$, applied to the spectral attention values $x_v$. This makes the estimated global illumination a multiplicative factor inside every layer's attention, which is what turns the network from a restorer into a calibrator.

What would settle it

Capture a scene with two clearly separated illumination conditions in one frame (shade and direct sun, or lamp plus daylight), build ground truth by placing a white reference in each region, and compare SIT's single output against both region-wise reflectance maps; if the model cannot reproduce both at once, the uniform-illumination assumption fails.

Watch

Extended reading notes

Core claim

The central claim is that global spectral illumination can be removed from a single hyperspectral image without a reference measurement, by learning the per-wavelength scale factors that turn observed intensity into reflectance under the model $I(\lambda)=R(\lambda)L(\lambda)$. The authors introduce the Spectral Illumination Transformer, a U-shaped transformer whose units combine spectral attention with an illumination attention branch that computes global per-channel averages after downsampling, mirroring Gray-World's assumption inside the network. The attention maps from both branches are multiplied and softmaxed to produce the final weights, and this design is what lets the network estimate and remove illumination instead of merely denoising. On the full-spectrum test set the method reaches 26.3 dB PSNR against 25.7 dB for the strongest baseline, and the gain is larger on the expanded dataset with synthetic illuminations.

Load-bearing premise

The paper assumes the same global spectrum of light reaches every pixel, so scenes with shadows, mixed light sources, or vignetting have no single per-band correction for the network to learn.

Editorial extensions

If this is right

  • Handheld or moving hyperspectral cameras could capture natural scenes once and still recover reflectance, removing the panel-occlusion and double-shot requirements of synchronous and asynchronous calibration.
  • The proposed 204-band, 400-1000 nm dataset and its 31-channel resampled version give the community a common training and evaluation resource for automatic calibration, with 765 real pairs expanded to 7,650 pairs under ten physically measured illuminations.
  • Because the illumination-attention branch drives most of the improvement, combining global illumination statistics with spectral attention is the promising design direction for future calibration networks.
  • Low-light regions and the near-infrared range are where every tested method loses accuracy, so further gains will likely come from illumination-adaptive or noise-aware calibration rather than architecture scaling alone.

Reading between the lines

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

  • The paper's uniform-global-illumination model means SIT is not tested on spatially varying light; I would expect errors to concentrate at shadow boundaries and vignetted edges, and the dataset's 'shadowy' condition, being one global spectrum applied to entire images, does not exercise this failure mode.
  • The illumination-attention branch computes a global spatial average per channel, so SIT should inherit Gray-World's vulnerability to scenes dominated by a single saturated color; a quick diagnostic would be to calibrate an image of a large red object under white light and check whether the output is tinted cyan.
  • A natural extension would be to condition the network on per-region or per-pixel illumination estimates and compare against the global SIT; if per-pixel conditioning closes the low-light gap, the uniform-illumination assumption is the real bottleneck.
Share X Bluesky LinkedIn Reddit HN

Signed reviews

No signed human review yet.

Editorial analysis

A structured set of objections, weighed in public.

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

Referee Report

3 major / 7 minor

Summary. The paper proposes automatic spectral calibration of hyperspectral images (HSI) without a physical reference panel. It introduces the BJTU-UVA dataset of 765 real uncalibrated/calibrated HSI pairs, captured with a 204-band 400-1000 nm camera using an asynchronous reference method, and an expanded BJTU-UVA-E dataset of 7650 pairs formed by multiplying scene reflectances with ten physically measured illuminant spectra. The authors also propose the Spectral Illumination Transformer (SIT), a U-shaped transformer with an additional illumination attention branch inspired by Gray-World. Experiments compare SIT against Gray-World, DivIll, SERT, and HCANet on the real and expanded benchmarks, with separate visible/NIR and per-illuminant analyses, and report state-of-the-art performance for SIT.

Significance. If the central claim holds, the paper provides a genuinely useful resource: the first public dataset specifically designed for automatic HSI calibration, with real paired data and an external test set. The real-data benchmark is not circular, which is a strength, and the illumination attention module is well motivated and shows a clear benefit in the ablation study (Table 3). The code and dataset availability, once clarified, would support reproducibility. However, the current quantitative evidence for the state-of-the-art claim is limited: on the real BJTU-UVA test set the margin over HCANet is only 0.6 dB PSNR and the SAM score is tied with Gray-World, while the large margins on the expanded benchmark are weakened by the benchmark's construction. The central method's assumption of spatially uniform global illumination is also not tested by any experiment.

major comments (3)
  1. [Sec. 4.2 and Sec. 6.1, Eq. (6), Tables 1 and 5] The BJTU-UVA-E benchmark is not a valid test of generalization to unseen illumination. The expanded images are generated as I_i^S = R_gt * L_i^S using only 10 fixed illuminant spectra, and the 5350/1140/1160 split over the 7650 pairs is described without any constraint that an illuminant or scene be excluded from training when it appears in test. Every one of the 10 illuminants thus appears in both training and test, and the same reflectance maps also recur under different illuminants. The large margins on this benchmark (e.g., PSNR 39.1 vs 37.3 for HCANet in Table 1; similar margins in Tables 4 and 5) may largely reflect interpolation over the 10 known spectra and per-scene correlations rather than automatic calibration under a genuinely unseen illumination. Please re-split the expanded set so that held-out illuminants (and, ideally, held-out scenes) are never seen in training, and report those results separately, or explicitly describe BJTU-UVA-E as an interpolation benchmark and do not use it as primary evidence for the state-of-the-art claim.
  2. [Sec. 6.2, Table 1] On the externally captured BJTU-UVA test set, the margin of SIT over HCANet is only 0.6 dB PSNR (26.3 vs 25.7) and the SAM is tied with Gray-World (both 3.1). No error bars, standard deviations, or significance tests are reported for any table. Because the margin is small, it is not clear that it is robust across random seeds, initializations, or the particular 116-image test split. Please report multiple training runs with variance and, if feasible, a paired statistical test over the test images. The state-of-the-art wording should be calibrated to the size of the margin on the real benchmark.
  3. [Sec. 3, Eqs. (1)-(3), (6)] The paper explicitly assumes spatially uniform global illumination: 'we focus on global illumination and assume it is uniform', and Eq. (1) uses a single per-band L(λ) for all pixels. The 'shadowy' illumination in Fig. 3 and Table 5 is still one global spectrum applied to entire images through Eq. (6), so none of the experiments expose the network to spatial illumination variation such as cast shadows, mixed light sources, or vignetting. Since the introduction motivates the method for natural scenes, the claims should either be restricted to uniform-illumination scenes or extended with a spatially varying test case (e.g., scenes with strong shadows compared against a per-pixel reference). Without this, the method's applicability to general natural scenes remains an untested assumption.
minor comments (7)
  1. [Abstract / Supp. Table 6] The availability statements are contradictory: the abstract and GitHub URL say the dataset and code are available online, while Supp. Table 6 states that the dataset 'will be publicly released in the future' and the license text says access is provided under the stated terms. Please state unambiguously which artifacts are available now, under which license, and provide a persistent link.
  2. [Sec. 4.1] The sentence beginning 'For caption, we use the asynchronous method' should read 'For capture' rather than 'For caption'.
  3. [Sec. 4.5, Eq. (9)] The RMSE formula uses the notation \(\hat{R}_{gt,i}\) which is confusing; it should compare the estimated reflectance \(\hat{R}_i\) with the ground truth \(R_{gt,i}\), not with a quantity named 'estimated ground truth'.
  4. [Sec. 4.3] The resampling procedure from 204 channels to 31 channels is not described. Please specify the spectral interpolation or binning method used and whether the same resampling was applied to all images consistently.
  5. [Throughout] The name of the Gray-World method is written inconsistently as 'GrayWorld', 'Gray-World', and 'Grayworld'; please unify the usage.
  6. [Fig. 3 and Table 5] The word 'puple' should be 'purple'.
  7. [Sec. 2.3] The text says 'Puria et al. proposed the DivIll [23]', but the reference is to Moghadam et al.; please use a consistent author citation.

Circularity Check

1 steps flagged · score 4.0 of 10

The BJTU-UVA-E expansion benchmark is partially circular: its test images are generated by the paper's Eq. (6) from the same 10 illuminant spectra and 765 reflectance maps used to create the training set, so the reported large margins partly reflect fitting known illuminants rather than generalizing to unseen illumination.

  1. fitted input called prediction [Sec. 4.2 (BJTU-UVA-E construction, Eq. 6); Sec. 6.1 (split into 5350/1140/1160); Table 1]
    "To enable further study of the illumination. We expand the dataset by combining the ground truth reflectance Rgt(λ) with 10 different physically measured illuminations LSi (λ). ISi (λ) =Rgt(λ)LSi (λ), i∈ 1, 2, · · ·10, (6) ... The BJTU-UV A dataset is split into three sets: 535 for training, 114 for validation, and 116 for testing. Similarly, for the expansion BJTU-UV A-E: 5350 for training, 1140 for validation, and 1160 for testing."

    Every BJTU-UVA-E image is the product of one of 765 scene reflectance maps and one of only 10 illuminant spectra, per Eq. (6). The paper describes only an image-count split and does not state that illuminants or scenes are disjoint between training and test. Therefore the same 10 LSi(λ) spectra that generate test inputs are present in training pairs (applied to other scenes), and the same Rgt(λ) maps are present under other illuminants. A network can memorize the 10 inverse spectra and solve the test set by recognizing which known illumination is present, so the large gains on BJTU-UVA-E (e.g., PSNR 39.1 vs 37.3 for HCANet in Table 1) are partly interpolation over known components rather than automatic calibration under genuinely unseen illumination.

full rationale

The paper's physical formation model (Eqs. 1-3), asynchronous ground truth (Eq. 5), and the SIT architecture are not circular: the real BJTU-UVA test set was captured independently with a physical reference, and the comparisons to Gray-World, DivIll, SERT, and HCANet are external. The ablation study is a normal within-model comparison. The only significant circularity is the synthetic expansion benchmark: because BJTU-UVA-E is generated by the paper's own Eq. (6) and the split is by image count rather than by illuminant or scene, the test set shares its generating components with the training set. This makes the expansion benchmark's large reported advantages partially a test of memorizing or interpolating the 10 known illuminant spectra, not of generalizing to unseen illumination. I do not count the small real-data margin or the absence of significance tests as circularity, and I note the contradictory release statements (abstract says 'available online' while Supp. Table 6 says 'will be publicly released in the future') as a reproducibility concern rather than a circularity concern.

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

The method is a supervised regressor; everything it knows is fit from the dataset. The physically load-bearing assumptions are uniform global illumination and the multiplicative image formation model, both stated, plus temporal stability of illumination during asynchronous capture, which is not quantified. No new physical entities are introduced.

free parameters (3)
  • SIT network weights = trained on BJTU-UVA and BJTU-UVA-E, not enumerated
    The central mapping f(I) is entirely learned; its weights are fit with L1 loss and Adam on the authors' dataset. Performance claims are tied to this fitted function, not to a parameter-free derivation.
  • IA down-scaling stride t and pooling size p = not specified
    Eqs. (14) and (15) define the illumination extractor with stride t and pooling size p, but the paper never gives their values, which matters for reproducing the method.
  • Number of SIT-U layers M and U-Net depth N = not specified
    Architecture depth is not reported; reproducibility requires these choices.
assumptions (3)
  • domain assumption Global illumination is uniform across the image (single L(λ) for all pixels).
    Stated in Sec. 3; used in Eq. (1) and in the asynchronous ground-truth computation Eq. (5). If false, the task is ill-posed.
  • domain assumption Image formation is pixel-wise multiplicative I(λ)=R(λ)L(λ) with no scattering, interreflection, or sensor nonlinearity beyond dark current.
    Sec. 3 Eq. (1). The synthetic expansion dataset Eq. (6) and ground truth Eq. (5) both rely on this model.
  • domain assumption Illumination is stable between the two asynchronous captures (L_{p,t1} ≈ L_{p,t2}) and dark current is additive and known.
    Sec. 4.1 Eq. (5). Ground truth quality depends on the time difference δ being small and on dark current subtraction being accurate.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Automatic Spectral Calibration of Hyperspectral Images:Method, Dataset and Benchmark." pith.science (2026). https://pith.science/paper/5KTOLOK6

@misc{pith2026241214925,
  author       = {Pith},
  title        = {Pith review of: Automatic Spectral Calibration of Hyperspectral Images:Method, Dataset and Benchmark},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/5KTOLOK6}},
  note         = {Machine review of arXiv:2412.14925}
}
read the original abstract

Hyperspectral image (HSI) densely samples the world in both the space and frequency domain and therefore is more distinctive than RGB images. Usually, HSI needs to be calibrated to minimize the impact of various illumination conditions. The traditional way to calibrate HSI utilizes a physical reference, which involves manual operations, occlusions, and/or limits camera mobility. These limitations inspire this paper to automatically calibrate HSIs using a learning-based method. Towards this goal, a large-scale HSI calibration dataset is created, which has 765 high-quality HSI pairs covering diversified natural scenes and illuminations. The dataset is further expanded to 7650 pairs by combining with 10 different physically measured illuminations. A spectral illumination transformer (SIT) together with an illumination attention module is proposed. Extensive benchmarks demonstrate the SoTA performance of the proposed SIT. The benchmarks also indicate that low-light conditions are more challenging than normal conditions. The dataset and codes are available online:https://github.com/duranze/Automatic-spectral-calibration-of-HSI

Figures

Figures reproduced from arXiv: 2412.14925 by the authors.

Figure 1
Figure 1. Comparison of HSI Calibration Methods: (a) Syn [PITH_FULL_IMAGE:figures/full_fig_p001_1.png] view at source ↗
Figure 2
Figure 2. Overview of the BJTU-UVA Dataset: (a) Dataset Recording Setup, (b) Variety of Scenes, (c) Illumination variety, and (d) HSI [PITH_FULL_IMAGE:figures/full_fig_p005_2.png] view at source ↗
Figure 3
Figure 3. Measurement of Ten Different Illuminations Using [PITH_FULL_IMAGE:figures/full_fig_p005_3.png] view at source ↗
Figures from the paper (7 more)
Figure 4
Figure 4. Figure 4: Structure of Spectral Illumination Transformer (SIT) framework and its unit (SIT-U) with Spectral and Illumination Attention [PITH_FULL_IMAGE:figures/full_fig_p006_4.png]
Figure 5
Figure 5. Figure 5: Visual Comparison of Absolute Error Using Heat Maps: The first two columns show the RGB rendering of the input and ground [PITH_FULL_IMAGE:figures/full_fig_p007_5.png]
Figure 6
Figure 6. Figure 6: Challenging Cases: Comparison of Calibrated Spectra [PITH_FULL_IMAGE:figures/full_fig_p008_6.png]
Figure 7
Figure 7. Figure 7: Visual Comparison of Absolute Error Using Heat Maps on BJTU-UVA. First row: Uncalibrated HSI, ground truth, and error [PITH_FULL_IMAGE:figures/full_fig_p015_7.png]
Figure 8
Figure 8. Figure 8: Visual Comparison of Absolute Error Using Heat Maps on BJTU-UVA. [PITH_FULL_IMAGE:figures/full_fig_p016_8.png]
Figure 9
Figure 9. Figure 9: Visual Comparison of Absolute Error Using Heat Maps on BJTU-UVA-E. [PITH_FULL_IMAGE:figures/full_fig_p017_9.png]
Figure 10
Figure 10. Figure 10: Visual Comparison of Absolute Error Using Heat Maps on BJTU-UVA-E. [PITH_FULL_IMAGE:figures/full_fig_p018_10.png]

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

47 extracted references · 45 canonical work pages

  1. [1]

    Auto white-balance correction for mixed-illuminant scenes

    Mahmoud Afifi, Marcus A Brubaker, and Michael S Brown. Auto white-balance correction for mixed-illuminant scenes. In Proceedings of the IEEE/CVF Winter Conference on Ap- plications of Computer Vision, pages 1210–1219, 2022. 2

    work page 2022

  2. [2]

    Sparse recovery of hyper- spectral signal from natural rgb images

    Boaz Arad and Ohad Ben-Shahar. Sparse recovery of hyper- spectral signal from natural rgb images. In European Con- ference on Computer Vision , pages 19–34. Springer, 2016. 1

    work page 2016

  3. [3]

    Ntire 2022 spectral recovery chal- lenge and data set

    Boaz Arad, Radu Timofte, Rony Yahel, Nimrod Morag, Amir Bernat, Yuanhao Cai, Jing Lin, Zudi Lin, Haoqian Wang, Yulun Zhang, et al. Ntire 2022 spectral recovery chal- lenge and data set. In Proceedings of the IEEE/CVF Con- ference on Computer Vision and Pattern Recognition, pages 863–881, 2022. 1, 6

    work page 2022

  4. [4]

    Character- ization of soil properties using reflectance spectroscopy

    E Ben-Dor, S Chabrillat, and Jos ´e AM Demattˆe. Character- ization of soil properties using reflectance spectroscopy. In Fundamentals, sensor systems, spectral libraries, and data mining for vegetation, pages 187–247. CRC press, 2018. 1, 3

    work page 2018

  5. [5]

    A spatial processor model for object colour perception

    Gershon Buchsbaum. A spatial processor model for object colour perception. Journal of the Franklin institute, 310(1): 1–26, 1980. 2, 5, 6, 8, 13

    work page 1980

  6. [6]

    Atomic spectrometry up- date: review of advances in the analysis of metals, chemicals and materials

    Simon Carter, Andy Fisher, Bridget Gibson, John Marshall, Ben Russell, and Ian Whiteside. Atomic spectrometry up- date: review of advances in the analysis of metals, chemicals and materials. Journal of Analytical Atomic Spectrometry , 32(11):2068–2117, 2017. 1, 3

    work page 2017

  7. [7]

    Statistics of real-world hyperspectral images

    Ayan Chakrabarti and Todd Zickler. Statistics of real-world hyperspectral images. In Proceedings of the IEEE/CVF Con- ference on Computer Vision and Pattern Recognition, pages 193–200. IEEE, 2011. 6

    work page 2011

  8. [8]

    Exploring the applica- bility of spectral recovery in semantic segmentation of rgb images

    Zhuoran Du, Shikui Wei, Ting Liu, Shunli Zhang, Xiaotong Chen, Shiyin Zhang, and Yao Zhao. Exploring the applica- bility of spectral recovery in semantic segmentation of rgb images. IEEE Transactions on Multimedia, 2023. 1, 6

    work page 2023

Show all 47 references
  1. [9]

    Image analysis operations applied to hyperspectral images for non-invasive sensing of food quality–a comprehensive review.Biosystems engineering, 142:53–82, 2016

    Gamal M ElMasry and Shigeki Nakauchi. Image analysis operations applied to hyperspectral images for non-invasive sensing of food quality–a comprehensive review.Biosystems engineering, 142:53–82, 2016. 3

    2016

  2. [10]

    Two-stage deep learning framework for srgb image white balance

    Marwa Farghaly, Romany F Mansour, and Adel A Sewisy. Two-stage deep learning framework for srgb image white balance. Signal, Image and Video Processing , 17(1):277– 284, 2023. 2

    2023

  3. [11]

    LG Ferreira, H Yoshioka, A Huete, and EE Sano. Opti- cal characterization of the brazilian savanna physiognomies for improved land cover monitoring of the cerrado biome: preliminary assessments from an airborne campaign over an lba core site. Journal of Arid Environments, 56(3...

  4. [12]

    Digital Image Processing

    Rafael C Gonzalez and Richard E Woods. Digital Image Processing. Prentice Hall, 2nd edition, 2002. 4

    2002

  5. [13]

    Hybrid convolutional and attention network for hyper- spectral image denoising

    Shuai Hu, Feng Gao, Xiaowei Zhou, Junyu Dong, and Qian Du. Hybrid convolutional and attention network for hyper- spectral image denoising. IEEE Geoscience and Remote Sensing Letters, 2024. 3, 4, 5, 6, 8, 13

    2024

  6. [14]

    Deterministic neural illumination mapping for efficient auto- white balance correction

    Furkan Kınlı, Do ˘ga Yılmaz, Barıs ¸¨Ozcan, and Furkan Kırac ¸. Deterministic neural illumination mapping for efficient auto- white balance correction. In Proceedings of the IEEE/CVF International Conference on Computer Vision, pages 1139– 1147, 2023. 2

    2023

  7. [15]

    Modeling the lighting in scenes as style for auto white- balance correction

    Furkan Kınlı, Do ˘ga Yılmaz, Barıs ¸¨Ozcan, and Furkan Kırac ¸. Modeling the lighting in scenes as style for auto white- balance correction. In Proceedings of the IEEE/CVF Win- ter Conference on Applications of Computer Vision , pages 4903–4913, 2023. 2

    2023

  8. [16]

    Goniometric measurements of a spray-painted barium sulfate reference panel

    George A Leshkevich. Goniometric measurements of a spray-painted barium sulfate reference panel. Remote Sens- ing of Environment, 24(2):287–296, 1988. 1, 3

    1988

  9. [17]

    Spec- tral mvir: Joint reconstruction of 3d shape and spectral re- flectance

    Chunyu Li, Yusuke Monno, and Masatoshi Okutomi. Spec- tral mvir: Joint reconstruction of 3d shape and spectral re- flectance. In 2021 IEEE International Conference on Com- putational Photography (ICCP), pages 1–12. IEEE, 2021. 1, 3

    2021

  10. [18]

    Swbnet: a stable white balance network for srgb im- ages

    Chunxiao Li, Xuejing Kang, Zhifeng Zhang, and Anlong Ming. Swbnet: a stable white balance network for srgb im- ages. In Proceedings of the AAAI Conference on Artificial Intelligence, pages 1278–1286, 2023. 2

    2023

  11. [19]

    Spectral enhanced rectangle transformer for hyperspectral image denoising

    Miaoyu Li, Ji Liu, Ying Fu, Yulun Zhang, and Dejing Dou. Spectral enhanced rectangle transformer for hyperspectral image denoising. In Proceedings of the IEEE/CVF Con- ference on Computer Vision and Pattern Recognition, pages 5805–5814, 2023. 3, 6, 8, 13, 14

    2023

  12. [20]

    Hyperspectral image denoising based on the similar spectra approaching

    Shouzhi Li, Liangliang Zhu, Luyan Ji, Xin Yi, Yongchao Zhao, and Xiurui Geng. Hyperspectral image denoising based on the similar spectra approaching. Infrared Physics & Technology, 131:104667, 2023. 3

    2023

  13. [21]

    Spectroscopic ellipsometry and polarimetry for materials and systems analysis at the nanometer scale: state-of-the-art, potential, and perspectives

    Maria Losurdo, Michael Bergmair, Giovanni Bruno, De- nis Cattelan, Christoph Cobet, Antonello De Martino, Karsten Fleischer, Zorana Dohcevic-Mitrovic, Norbert Esser, Melanie Galliet, et al. Spectroscopic ellipsometry and polarimetry for materials and systems analysis at the na...

    2009

  14. [22]

    A technique for re- flectance calibration of airborne hyperspectral spectrometer data using a broad, multiband radiometer

    Tomoaki Miura, Alfredo R Huete, Laerte Guima˜aes Ferreira, Edson E Sano, and Hiroki Yoshioka. A technique for re- flectance calibration of airborne hyperspectral spectrometer data using a broad, multiband radiometer. In Hyperspectral Remote Sensing of Tropical and Sub-tropical...

    2008

  15. [23]

    Enabling hyperspectral imaging in diverse illumi- nation conditions for indoor applications

    Puria Azadi Moghadam, Neha Sharma, and Mohamed Hefeeda. Enabling hyperspectral imaging in diverse illumi- nation conditions for indoor applications. In Proceedings of the 12th ACM Multimedia Systems Conference, pages 23–35,

  16. [24]

    Mul- tiscale adaptive fusion network for hyperspectral image de- noising

    Haodong Pan, Feng Gao, Junyu Dong, and Qian Du. Mul- tiscale adaptive fusion network for hyperspectral image de- noising. IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, 16:3045–3059, 2023. 3

    2023

  17. [25]

    Angular and seasonal variation of spectral surface reflectance ratios: Implications for the remote sensing of aerosol over 10 land

    Lorraine A Remer, Andrew E Wald, and Yoram J Kaufman. Angular and seasonal variation of spectral surface reflectance ratios: Implications for the remote sensing of aerosol over 10 land. IEEE Transactions on Geoscience and Remote Sens- ing, 39(2):275–283, 2001. 2

    2001

  18. [26]

    U- net: Convolutional networks for biomedical image segmen- tation

    Olaf Ronneberger, Philipp Fischer, and Thomas Brox. U- net: Convolutional networks for biomedical image segmen- tation. In Medical image computing and computer-assisted intervention–MICCAI 2015: 18th international conference, Munich, Germany, October 5-9, 2015, proceedings, pa...

    2015

  19. [27]

    A new approach for the analysis of hyperspectral data: Theory and sensitivity analysis of the moment distance method

    Eric Ariel L Salas and Geoffrey M Henebry. A new approach for the analysis of hyperspectral data: Theory and sensitivity analysis of the moment distance method. Remote sensing, 6 (1):20–41, 2013. 3

    2013

  20. [28]

    Specim - Hyperspectral Imaging Cameras and Systems

    Specim, Spectral Imaging Ltd. Specim - Hyperspectral Imaging Cameras and Systems. https://www.specim. com/. 3

  21. [29]

    Quality of high resolution synthesized images: Is there a simple criterion? In Proceedings of the Third Con- ference ”Fusion of Earth Data”, pages 99–103, 2000

    Lucien Wald. Quality of high resolution synthesized images: Is there a simple criterion? In Proceedings of the Third Con- ference ”Fusion of Earth Data”, pages 99–103, 2000. 4

    2000

  22. [30]

    Image quality assessment: From error visibility to structural similarity

    Zhou Wang, Alan C Bovik, Hamid R Sheikh, and Eero P Si- moncelli. Image quality assessment: From error visibility to structural similarity. IEEE Transactions on Image Process- ing, 13(4):600–612, 2004. 4

    2004

  23. [31]

    Uformer: A general u-shaped transformer for image restoration

    Zhendong Wang, Xiaodong Cun, Jianmin Bao, Wengang Zhou, Jianzhuang Liu, and Houqiang Li. Uformer: A general u-shaped transformer for image restoration. In Proceedings of the IEEE/CVF conference on computer vision and pattern recognition, pages 17683–17693, 2022. 4

    2022

  24. [32]

    Generalized assorted pixel camera: post- capture control of resolution, dynamic range, and spectrum

    Fumihito Yasuma, Tomoo Mitsunaga, Daisuke Iso, and Shree K Nayar. Generalized assorted pixel camera: post- capture control of resolution, dynamic range, and spectrum. IEEE transactions on image processing , 19(9):2241–2253,

  25. [33]

    Discrimination among semi-arid landscape endmem- bers using the spectral angle mapper (sam) algorithm

    Richard H Yuhas, Alexander FH Goetz, and Joe W Board- man. Discrimination among semi-arid landscape endmem- bers using the spectral angle mapper (sam) algorithm. In Summaries of the Third Annual JPL Airborne Geoscience Workshop, pages 147–149, 1992. 4

    1992

  26. [34]

    Hyperspectral image de- noising: From model-driven, data-driven, to model-data- driven

    Qiang Zhang, Yaming Zheng, Qiangqiang Yuan, Meiping Song, Haoyang Yu, and Yi Xiao. Hyperspectral image de- noising: From model-driven, data-driven, to model-data- driven. IEEE Transactions on Neural Networks and Learn- ing Systems, 2023. 3 11 Automatic Spectral Calibration of ...

    2023

  27. [35]

    We conduct experiments on both the BJTU-UV A and BJTU-UV A-E datasets

    Detailed Evaluation of Calibration Methods: Visible and Near-Infrared Ranges To further verify the quality of the spectral calibration, we evaluate the calibrated HSIs from the visible (400-700nm) and near-infrared (700-1000nm) perspectives. We conduct experiments on both the ...

  28. [36]

    The definitions and spec- tra of these illumination settings are illustrated in Fig

    Evaluation on Different Illumination Condi- tions of the BJTU-UV A-E Dataset To evaluate model performance under varying illumination conditions, we analyzed the PSNR calibration results across ten different illumination settings. The definitions and spec- tra of these illumin...

  29. [37]

    7 and Fig

    More Qualitative Analysis We present six examples from the BJTU-UV A dataset in Fig. 7 and Fig. 8, and six additional examples in Fig. 9 and Fig. 10. Each example includes the following evaluations: • First row: The first two images represent the uncali- brated HSI and the gro...

  30. [38]

    Comparison of public hyperspectral image datasets

    Comparison with Existing Hyperspectral Datasets for Natural Scenes As shown in Table 6, the rapid development of hyper- spectral imaging devices has shifted research focus from 13 Dataset Bands Images Spectral Range Spatial Resolution Denoising Super-resolution Spectral Recove...

    2010

  31. [39]

    Fumihito Yasuma, Tomoo Mitsunaga, Daisuke Iso, and Shree K. Nayar. Generalized assorted pixel camera: post- capture control of resolution, dynamic range, and spectrum. IEEE Transactions on Image Processing, 19(9):2241–2253, 2010

    2010

  32. [40]

    Statistics of real-world hyperspectral images

    Ayan Chakrabarti and Todd Zickler. Statistics of real-world hyperspectral images. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pages 193–200, IEEE, 2011

    2011

  33. [41]

    Prasad, and Michael S

    Rang MH Nguyen, Dilip K. Prasad, and Michael S. Brown. Training-based spectral reconstruction from a sin- gle RGB image. In European Conference on Computer Vi- sion, pages 186–201. Springer, 2014

    2014

  34. [42]

    Sparse recov- ery of hyperspectral signal from natural RGB images

    Boaz Arad and Ohad Ben-Shahar. Sparse recov- ery of hyperspectral signal from natural RGB images. In European Conference on Computer Vision, pages 19–34. Springer, 2016

    2016

  35. [43]

    NTIRE 2018 challenge on spectral reconstruction from RGB images

    Boaz Arad, Ohad Ben-Shahar, and Radu Timofte. NTIRE 2018 challenge on spectral reconstruction from RGB images. In Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition Workshops, pages 929–938, 2018

    2018

  36. [44]

    Finlayson

    Boaz Arad, Radu Timofte, Ohad Ben-Shahar, Yi-Tun Lin, and Graham D. Finlayson. NTIRE 2020 challenge on spectral reconstruction from an RGB image. In Proceed- ings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops, pages 446–447, 2020

    2020

  37. [45]

    NTIRE 2022 spectral recov- ery challenge and dataset

    Boaz Arad, Radu Timofte, Rony Yahel, Nimrod Morag, Amir Bernat, Yuanhao Cai, Jing Lin, Zudi Lin, Hao- qian Wang, Yulun Zhang, et al. NTIRE 2022 spectral recov- ery challenge and dataset. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, page...

    2022

  38. [46]

    Exploring the applicabil- ity of spectral recovery in semantic segmentation of RGB images

    Du Z., Wei S., Liu T., et al. Exploring the applicabil- ity of spectral recovery in semantic segmentation of RGB images. IEEE Transactions on Multimedia, 2023

    2023

  39. [47]

    ES- SAformer: Efficient transformer for hyperspectral im- age super-resolution

    Zhang M., Zhang C., Zhang Q., et al. ES- SAformer: Efficient transformer for hyperspectral im- age super-resolution. In Proceedings of the IEEE/CVF International Conference on Computer Vision, pages 23073–23084, 2023. 14 Gray-WorldUncalibrated GT HCANet DivIll SERT SIT (a) (b)...

    2023

Pith tools

Reviewed August 11, 2026 · model on record in the stance chip above.