Pith. sign in

REVIEW 3 major objections 4 minor 1 cited by

FLAME 3 Dataset: Unleashing the Power of Radiometric Thermal UAV Imagery for Wildfire Management

T0 review · 3 major / 4 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 A new drone dataset pairs visible video with per-pixel temperature maps from prescribed fires, and models trained on the raw temperatures outperform RGB-only fire detectors.

desk verdict Valuable new data resource with a circular benchmark that needs fixing before the performance claims can be taken seriously. read the letter →

arxiv 2412.02831 v2 pith:ZMM3FCKM submitted 2024-12-03 cs.CV cs.AI

Bryce Hopkins , Leo ONeill , Michael Marinaccio , Mobin Habibpour , Eric Rowell , Russell Parsons , Sarah Flanary , Irtija Nazim
show 2 more authors
Carl Seielstad Fatemeh Afghah
This is my paper · ORCID
classification cs.CVcs.AI
keywords wildfiredetectionradiometricthermalimageryUAVTIFFRGB-thermalpairsprescribedfireimagedatasetprocessingpipeline
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

This paper argues that the missing ingredient in AI wildfire management is radiometric thermal data: per-pixel temperature values rather than color-mapped infrared images. It presents FLAME 3, a dataset of synchronized visible and radiometric long-wave infrared imagery collected by drones at six prescribed burns, together with a processing pipeline that turns raw drone files into aligned RGB/thermal-TIFF pairs ready for machine learning. The paper claims this is the first comprehensive aerial dual-spectrum radiometric fire dataset, and reports that models trained on the temperature TIFFs reach about 91% Fire/No-Fire test accuracy, outperforming RGB-only and JPEG-thermal inputs and generalizing across burn sites. If correct, the dataset removes the main obstacle to temperature-aware wildfire detection, segmentation, and fire-behavior modeling.

What carries the argument

The load-bearing object is the radiometric thermal TIFF: a one-band raster in which every pixel stores a temperature estimate from the long-wave infrared camera. It carries the argument because temperature is the quantity that separates fire from non-fire directly, which lets the authors auto-generate Fire/No-Fire labels by thresholding, gives pixel-level temperature ground truth for possible segmentation and regression tasks, and provides the input modality that scores highest in their classification tests. The supporting machinery is the FOV-correction step that crops and scales RGB frames toward the thermal camera's 640x512 view so that paired data can later be fused by convolution.

What would settle it

Have expert annotators label a held-out set of FLAME 3 frames as Fire/No-Fire using visible evidence and fire perimeters, ignoring the TIFF temperature values, then retrain and test the TIFF-only model on those labels; if its accuracy drops to the level of RGB-only models, the claim that radiometric inputs 'generalize much better' is not supported.

Watch

Extended reading notes

Core claim

FLAME 3 provides paired RGB images and radiometric thermal TIFF files from six prescribed burns in pine, grass, and sagebrush fuels, where each TIFF pixel is a temperature reading rather than a color. The authors' extraction pipeline converts proprietary radiometric JPEG metadata into standard TIFFs, regenerates thermal JPEGs with a fixed color map, and crops/scales the RGB frames to approximate 1:1 pixel alignment with the thermal camera. Temperature thresholds automatically assign preliminary Fire/No-Fire labels (below about 80°C as No-Fire, above about 200°C as Fire), which human reviewers then clean. In their classification comparison, the FLAME 3 TIFF-only input achieved 91.38% test accuracy and RGB-TIFF 90.95%, above all FLAME 1/FLAME 2 combinations, which the authors attribute to the richer contextual information in radiometric data. The dataset also includes nadir thermal plots, repeat overhead radiometric sequences that can measure rate of spread and energy release.

Load-bearing premise

The Fire/No-Fire labels used to judge the models come from temperature thresholds applied to the very same TIFF pixels the best model reads as input, so the headline accuracy partly measures the threshold rather than independent ground truth.

Editorial extensions

If this is right

  • Models trained on FLAME 3 can treat temperature as a first-class input, making fire visible through smoke and reducing false positives that plague RGB-only detectors.
  • The same TIFFs can auto-generate pixel-level Fire/No-Fire masks via Otsu or hysteresis thresholding, cutting the manual labeling burden for segmentation.
  • Nadir thermal plots provide repeat georeferenced temperature maps that can support rate-of-spread and energy-release measurements during prescribed burns.
  • The classification results imply that replacing JPEG thermal images with radiometric TIFFs improves accuracy even with fewer training samples.
  • The published collection and processing procedures give other groups a repeatable path to gather radiometric dual-spectrum fire data.

Reading between the lines

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

  • Because the benchmark labels are created by temperature thresholds on the same TIFFs the top model sees, the 91% accuracy likely overstates real-world generalization; an independent label source, such as expert-annotated fire perimeters, would give a fairer comparison.
  • The fully aligned pixel pairs could enable a model to regress temperature from RGB alone, letting non-radiometric cameras act as cheap temperature proxies during future fires.
  • The six-burn diversity is also a stress test: if the dataset is used as a pretraining pool, held-out wildfires (not prescribed burns) should be used to see whether radiometric features transfer to more intense fire behavior.
  • Reducing the residual ~20-pixel alignment error with algorithmic distortion correction would unlock early convolutional fusion of RGB and thermal layers, which the authors expect to outperform late fusion.
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 / 4 minor

Summary. This manuscript introduces FLAME 3, a UAV-collected dataset of side-by-side RGB and radiometric long-wave infrared imagery from six prescribed fires, together with thermal TIFF files and nadir thermal plot imagery. It also presents a semi-automated processing pipeline that pairs images, extracts radiometric TIFFs, and applies manual field-of-view corrections, and it reports classification experiments comparing FLAME 1, FLAME 2, and FLAME 3 under RGB-only, RGB-thermal JPEG, RGB-TIFF, and TIFF-only inputs. The authors report that FLAME 3 TIFF-only input achieves the highest test accuracy (91.38% on a Sycan Marsh test set) and claim this demonstrates that radiometric TIFF data provide richer contextual information and better generalization.

Significance. If the dataset is made publicly available as described, it would fill a genuine gap: publicly accessible aerial radiometric thermal wildfire imagery is currently almost nonexistent, and the per-pixel temperature information could support detection, segmentation, and fire-behavior modeling. The data collection guidelines and the processing pipeline are practical contributions, and the Kaggle subset provides a checkable artifact. However, the classification evaluation that supports the paper's headline claim about the value of TIFF data is confounded by the fact that the Fire/No-Fire labels are derived from temperature thresholds in the same TIFF modality, and the comparison lacks statistical rigor. These issues do not invalidate the dataset itself, but they require revision of the empirical claims before the paper can be accepted.

major comments (3)
  1. [Section III.A.1 and Section V.B (Table VIII)] The Fire/No-Fire labels are generated by applying temperature thresholds to the radiometric TIFF data: images with maximum temperature below approximately 80°C are labeled No-Fire, those above approximately 200°C are labeled Fire, with manual review only for intermediate cases. The TIFF-only classifier takes exactly the same temperature values as input. An accuracy of 91.38% on the Sycan test set can therefore largely be achieved by learning the labeling rule itself, and it does not by itself demonstrate that TIFF files 'contain rich contextual information' or 'generalize much better' across environments. To support the central quantitative claim, please either evaluate TIFF inputs against labels generated independently of the temperature data (for example, manual annotation from RGB imagery or independent human review of the borderline cases), or report a simple baseline that classifies each image by its maximum temperature using the same thresholds and show that the learned model substantially exceeds that baseline. Without this control, Table VIII cannot support the claim that radiometric TIFF inputs improve wildfire classification.
  2. [Section V.A (Table VII)] The FLAME 3 training set contains imagery from Shoetank, and the FLAME 3 test set also contains imagery from Shoetank. The text states that the specific testing images from Shoetank were 'randomly sampled and removed from the training set,' but other Shoetank images remain in training. This same-burn overlap means that the reported FLAME 3 test accuracy may be optimistically biased and does not demonstrate generalization to unseen burn sites or fuel types. Please either exclude the entire Shoetank burn from training when evaluating on Shoetank, or report per-burn results so that readers can assess cross-burn performance. The exact construction of the train/test split should also be stated unambiguously.
  3. [Section V (Tables VII and VIII)] All classification results are reported as single point estimates without error bars, confidence intervals, or the number of training runs per configuration. With 232 test images in Table VIII, the 0.43 percentage point difference between TIFF-only (91.38%) and RGB-TIFF (90.95%) is well within sampling noise, and the ranking of the top-performing inputs is not statistically supported. Please report means and standard deviations over multiple seeds, or at least binomial confidence intervals for the test accuracies, and moderate the comparative claims accordingly.
minor comments (4)
  1. [Section V.A] There is a typo in the text: 'hyperaparameter' should be 'hyperparameter'.
  2. [Section IV.B.1] 'Eucledian distance' should be 'Euclidean distance'.
  3. [Abstract and Section II.B.2] The novelty claim is phrased differently in the abstract ('first comprehensive collection') and in Section II.B.2 ('first ever aerial radiometric dual multi-spectral UAV-collected prescribed fire imagery dataset'); please align these wordings to avoid ambiguity.
  4. [Figure 8] The caption of Figure 8 says 'from Figure 8' where it appears to mean the overlay shown in Figure 7; please correct the cross-reference.

Circularity Check

1 steps flagged · score 7.0 of 10

TIFF classification benchmark is circular: Fire/No-Fire labels derive from temperature thresholds in the same TIFF inputs, so the 91.38% accuracy claim does not demonstrate generalization.

  1. self definitional [Section III.A.1 (threshold labeling) applied in Section V.B / Table VIII (TIFF evaluation)]
    "By applying thresholding techniques to the radiometric data, it effectively distinguishes between obvious Fire and No-Fire images. Specifically, images with a maximum temperature below approximately 80°C are highly likely to be classified as No-Fire, while those with temperatures exceeding 200°C typically depict active fire. ... The FLAME 3 testing set contained 116 Fire and 116 No Fire images from Sycan Marsh."

    The Fire/No-Fire labels used as ground truth in the TIFF classification experiment are generated by thresholding the same radiometric temperature values that the TIFF-only input provides. A model can therefore 'predict' the label by applying the same temperature cutoff, so the reported accuracy mostly confirms the labeling rule rather than demonstrating that radiometric TIFFs carry richer fire semantics or generalize across environments. Manual review of intermediate-temperature images does not remove the overlap for the threshold-defined extreme cases that populate the benchmark.

full rationale

The paper's main contribution is a new dataset and processing pipeline, which are independent of the circular benchmark. The RGB-only unimodal evaluation (Section V.A) compares FLAME 3 against FLAME 1/2 using externally supplied labels from those prior datasets, and that comparison is not circular. The circularity is confined to Section V.B, where the headline claim that TIFF inputs 'generalize much better' is evaluated against Fire/No-Fire labels that were themselves produced by thresholding the radiometric temperature data contained in the TIFF inputs. Because an image whose maximum temperature exceeds ~200°C is labeled Fire and one below ~80°C is labeled No-Fire, a model reading the same temperature values can achieve high accuracy by rediscovering the labeling threshold; the 91.38% TIFF-only accuracy and the claim of 'rich contextual information' are therefore partly restatements of the labeling rule. The paper is transparent about the threshold-based preliminary labeling, but it does not provide a control (e.g., labels from visible-spectrum human annotation or from an independent burn map) that would separate genuine generalization from label-input overlap. The dataset itself remains valuable, but this particular quantitative argument is not secure.

Assumptions & free parameters 2 free parameters · 4 assumptions · 0 invented entities

The central claims rest on the integrity of the collected imagery, the accuracy of radiometric temperature extraction, the validity of threshold-based labels, and the adequacy of manual FOV alignment. No new physical entities are introduced.

free parameters (2)
  • Temperature thresholds for Fire/No-Fire labeling = 80°C lower bound, 200°C upper bound
    Ad hoc thresholds chosen to auto-label images in Section III.A.1. They become the ground truth for the classifier evaluation, and the same temperature data is used as input, so the benchmark result is partly determined by these thresholds.
  • Manual FOV correction parameters (crop, scale, translation) = Not reported; tuned per camera and burn
    Section III.A.2 and Section IV.B.1 describe iterative adjustment to minimize pixel error. The exact parameters are not published, which affects reproducibility and the claimed alignment quality.
assumptions (4)
  • domain assumption Radiometric rJPEG metadata can be decoded into accurate per-pixel temperature values by the provided script.
    Section IV.B.1 extracts temperatures from raw thermal JPG metadata; no independent calibration check against a reference source is reported.
  • domain assumption Prescribed fires at six sites represent wildland fire conditions sufficiently for training detection models.
    Section II admits prescribed fires are low intensity and limited in behavior range, yet the dataset is positioned for wildfire management use.
  • domain assumption Manual FOV correction gives adequate RGB-thermal pixel correspondence for paired-image tasks.
    Section III.A.2 estimates maximum pixel error around 20 pixels after manual cropping and translation but does not validate task-level effects.
  • domain assumption Threshold-generated labels plus human review are an acceptable proxy for true Fire/No-Fire ground truth.
    Section III.A.1 uses 80°C and 200°C thresholds for initial labeling; correctness of the final labels is not separately measured.

how reviews work

0 comments
Cite this review

Pith. "Pith review of FLAME 3 Dataset: Unleashing the Power of Radiometric Thermal UAV Imagery for Wildfire Management." pith.science (2026). https://pith.science/paper/ZMM3FCKM

@misc{pith2026241202831,
  author       = {Pith},
  title        = {Pith review of: FLAME 3 Dataset: Unleashing the Power of Radiometric Thermal UAV Imagery for Wildfire Management},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/ZMM3FCKM}},
  note         = {Machine review of arXiv:2412.02831}
}
read the original abstract

The increasing accessibility of radiometric thermal imaging sensors for unmanned aerial vehicles (UAVs) offers significant potential for advancing AI-driven aerial wildfire management. Radiometric imaging provides per-pixel temperature estimates, a valuable improvement over non-radiometric data that requires irradiance measurements to be converted into visible images using RGB color palettes. Despite its benefits, this technology has been underutilized largely due to a lack of available data for researchers. This study addresses this gap by introducing methods for collecting and processing synchronized visual spectrum and radiometric thermal imagery using UAVs at prescribed fires. The included imagery processing pipeline drastically simplifies and partially automates each step from data collection to neural network input. Further, we present the FLAME 3 dataset, the first comprehensive collection of side-by-side visual spectrum and radiometric thermal imagery of wildland fires. Building on our previous FLAME 1 and FLAME 2 datasets, FLAME 3 includes radiometric thermal Tag Image File Format (TIFFs) and nadir thermal plots, providing a new data type and collection method. This dataset aims to spur a new generation of machine learning models utilizing radiometric thermal imagery, potentially trivializing tasks such as aerial wildfire detection, segmentation, and assessment. A single-burn subset of FLAME 3 for computer vision applications is available on Kaggle with the full 6 burn set available to readers upon request.

Figures

Figures reproduced from arXiv: 2412.02831 by the authors.

Figure 1
Figure 1. Three Images from Willamette prescribed burn on September 23rd, [PITH_FULL_IMAGE:figures/full_fig_p004_1.png] view at source ↗
Figure 2
Figure 2. RGB-IR Image Pairs from Hanna Hammock prescribed burn. From [PITH_FULL_IMAGE:figures/full_fig_p006_2.png] view at source ↗
Figure 3
Figure 3. Raw RGB-IR image Pair collected from each prescribed burn site [PITH_FULL_IMAGE:figures/full_fig_p006_3.png] view at source ↗
Figures from the paper (4 more)
Figure 5
Figure 5. Figure 5: FOV Corrections Process: Raw RGB Image (Input), FOV Corrected [PITH_FULL_IMAGE:figures/full_fig_p009_5.png]
Figure 4
Figure 4. Figure 4: TIFF Creation Process: Raw Thermal RJPEG Image (Input), Thermal [PITH_FULL_IMAGE:figures/full_fig_p009_4.png]
Figure 7
Figure 7. Figure 7: Overlaying of processed RGB-IR image pairs from Willamette (Top [PITH_FULL_IMAGE:figures/full_fig_p010_7.png]
Figure 8
Figure 8. Figure 8: Error in RGB-IR image pair alignment associated with distortion from [PITH_FULL_IMAGE:figures/full_fig_p010_8.png]

Discussion (0). Continue with ORCID to comment.

Forward citations

Cited by 1 Pith paper

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score. Full citation record

  1. Seeing Heat with Color -- RGB-Only Wildfire Temperature Inference from SAM-Guided Multimodal Distillation using Radiometric Ground Truth

    eess.IV 2025-05 conditional novelty 6.0 of 10

    SAM-TIFF trains an RGB-only model to segment fire and estimate per-pixel temperatures, reaching about 68% of fire pixels within 25°C of radiometric ground truth on FLAME 3.

Reference graph

Works this paper leans on

56 extracted references · 51 canonical work pages · cited by 1 Pith paper

  1. [1]

    Flame 3 computer vision subset (sycan marsh),

    B. Hopkins, L. O’Neill, M. Marinaccio, F. Afghah, E. Rowell, R. Parsons, and S. Flanary, “Flame 3 computer vision subset (sycan marsh),” 2024. [Online]. Available: https://www.kaggle.com/dsv/ 8480870

    work page 2024

  2. [2]

    Flame 3 -nadir thermal fire image dataset,

    B. Hopkins, L. O’Neill, M. Marinaccio, F. Afghah, E. Rowell, and R. Parsons, “Flame 3 -nadir thermal fire image dataset,” 2024. [Online]. Available: https://www.kaggle.com/dsv/8724543

    work page arXiv 2024

  3. [3]

    A comprehensive survey of research towards ai-enabled unmanned aerial systems in pre-, active-, and post-wildfire management,

    S. P. H. Boroujeni, A. Razi, S. Khoshdel et al., “A comprehensive survey of research towards ai-enabled unmanned aerial systems in pre-, active-, and post-wildfire management,” Information Fusion, vol. 108, p. 102369, 2024

    work page 2024

  4. [4]

    UA VS-FDDB: UA Vs- based Forest Fire Detection Database,

    M. N. Mowla, D. Asadi, and K. N. Tekeoglu, “UA VS-FDDB: UA Vs- based Forest Fire Detection Database,” 2024

    work page 2024

  5. [5]

    A rgb-thermal based adaptive modality learning network for day–night wildfire identification,

    X. Rui, Z. Li, X. Zhang et al., “A rgb-thermal based adaptive modality learning network for day–night wildfire identification,” International Journal of Applied Earth Observation and Geoinformation , vol. 125, p. 103554, 2023

    work page 2023

  6. [6]

    RGB-thermal wildfire dataset,

    ——, “RGB-thermal wildfire dataset,” 2022. [Online]. Available: https://rec.ustc.edu.cn/share/75aae550-8d26-11ee-98a1-ed4fd857523e

    work page 2022

  7. [7]

    Fire and smoke dataset,

    DataCluster Labs, “Fire and smoke dataset,” 2021. [Online]. Available: https://www.kaggle.com/datasets/dataclusterlabs/fire-and-smoke-dataset

    work page 2021

  8. [8]

    Wildfire detection image data,

    Dincer, Baris, “Wildfire detection image data,” 2021. [Online]. Available: https://www.kaggle.com/datasets/brsdincer/ wildfire-detection-image-data?resource=download

    work page 2021

Show all 56 references
  1. [9]

    Emergencynet: Efficient aerial image classification for drone-based emergency monitoring using atrous con- volutional feature fusion,

    C. Kyrkou and T. Theocharides, “Emergencynet: Efficient aerial image classification for drone-based emergency monitoring using atrous con- volutional feature fusion,” IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing , vol. 13, pp. 1687–1699, 2020

    2020

  2. [10]

    Deep-learning- based aerial image classification for emergency response applications using unmanned aerial vehicles,

    Kyrkou, Christos and Theocharides, Theocharis, “Deep-learning- based aerial image classification for emergency response applications using unmanned aerial vehicles,” 2019. [Online]. Available: https: //arxiv.org/abs/1906.08716

    2019 arXiv

  3. [11]

    Dataset for forest fire detection,

    A. Khan and B. Hassan, “Dataset for forest fire detection,” August 2020. [Online]. Available: https://data.mendeley.com/datasets/gjmr63rz2r/1

    2020

  4. [12]

    Fire detection,

    D. Shah, “Fire detection,” July 2020. [Online]. Available: https: //github.com/jackfrost1411/fire-detection

    2020

  5. [13]

    FireNET,

    M. Olafenwa, “FireNET,” August 2019. [Online]. Available: https: //github.com/OlafenwaMoses/FireNET

    2019

  6. [14]

    Fire Detection from CCTV,

    R. Pande, “Fire Detection from CCTV,” 2019. [Online]. Available: https://www.kaggle.com/datasets/ritupande/fire-detection-from-cctv

    2019

  7. [15]

    Experimental exploration of compact convolutional neural network architectures for non-temporal real-time fire detection,

    G. Samarth, N. Bhowmik, and T. Breckon, “Experimental exploration of compact convolutional neural network architectures for non-temporal real-time fire detection,” in Proc. International Conference on Machine Learning Applications. IEEE, December 2019, pp. 653–658

    2019

  8. [16]

    Experimentally defined convolutional nerual network architecture variants for non-temporal real-time fire detection,

    A. Dunnings and T. Breckon, “Experimentally defined convolutional nerual network architecture variants for non-temporal real-time fire detection,” in Proc. International Conference on Image Processing . IEEE, September 2018, pp. 1558–1562

    2018

  9. [17]

    Fire detection image dataset,

    J. Sharma and M. Goodwin, “Fire detection image dataset,” 2017. [On- line]. Available: https://github.com/cair/Fire-Detection-Image-Dataset

    2017

  10. [18]

    Computer vision for wildfire research: An evolving image dataset for processing and analysis,

    T. Toulouse, L. Rossi, A. Campana et al., “Computer vision for wildfire research: An evolving image dataset for processing and analysis,” Fire Safety Journal, vol. 92, pp. 188–194, 2017. 13

    2017

  11. [19]

    Real-time fire detection for video surveillance applications using a combination of experts based on color, shape and motion,

    P. Foggia, A. Saggese, and M. Vento, “Real-time fire detection for video surveillance applications using a combination of experts based on color, shape and motion,” IEEE Transactions on Circuits and Systems for Video Technology, 2015

    2015

  12. [20]

    Improving fire detection reli- ability by a combination of videoanalytics,

    R. D. Lascio, A. Greco, A. Saggese et al., “Improving fire detection reli- ability by a combination of videoanalytics,” in International Conference on Image Analysis and Recognition (ICIAR) , 2014

    2014

  13. [21]

    Aerial imagery pile burn detection using deep learning: The flame dataset,

    A. Shamsoshoara, F. Afghah, A. Razi et al., “Aerial imagery pile burn detection using deep learning: The flame dataset,” Computer Networks, vol. 193, p. 108001, 2021

    2021

  14. [23]

    Shamsoshoara and F

    A. Shamsoshoara and F. Afghah, Airborne Fire Detection and Modeling Using Unmanned Aerial Vehicles Imagery: Datasets and Approaches . Cham: Springer International Publishing, 2023, pp. 525–550

    2023

  15. [24]

    Building a virtual reality fire environment based on fire dynamic simulator,

    M. Tzani, J. Besharat, V . Charalampous et al., “Building a virtual reality fire environment based on fire dynamic simulator,” in 2020 International Conference on Information Technologies (InfoTech) , 2020, pp. 1–5

    2020

  16. [25]

    Fluid dynamics of wildfires Once a wildfire is ignited, complex interactions with the local winds affect how it behaves

    R. Linn, “Fluid dynamics of wildfires Once a wildfire is ignited, complex interactions with the local winds affect how it behaves.” Physics Today, vol. 72, no. 11, pp. 70–70, 11 2019

    2019

  17. [26]

    Using a computational fluid dynamic model to guide wildland fire management,

    J. Furman, R. Linn, J. Winterkamp et al. , “Using a computational fluid dynamic model to guide wildland fire management,” EGLIN AIR FORCE, Technical Report AD1135449, 04 2019, [Technical Report, Final Report]

    2019

  18. [27]

    Real-time flame segmentation based on RGB-thermal fusion,

    S. Guo, B. Hu, and R. Huang, “Real-time flame segmentation based on RGB-thermal fusion,” in 2021 IEEE International Conference on Robotics and Biomimetics (ROBIO) , 2021, pp. 1435–1440

    2021

  19. [28]

    Wuunet: Advanced fully convo- lutional neural network for multiclass fire segmentation,

    V . S. Bochkov and L. Y . Kataeva, “Wuunet: Advanced fully convo- lutional neural network for multiclass fire segmentation,” Symmetry, vol. 13, no. 1, p. 98, 2021

    2021

  20. [29]

    A model for the simulation of forest fire dynamics using cellular automata,

    D. D’Ambrosio, S. Di Gregorio, W. Spataro et al. , “A model for the simulation of forest fire dynamics using cellular automata,” in A. Voinov, A. Jakeman et A. Rizzoli, ´editeurs: Proceedings of the iEMSs Third Biennial Meeting:” Summit on Environmental Modelling and Software , 2006

    2006

  21. [30]

    Learning-based prediction of wildfire spread with real-time rate of spread measurement,

    C. Zhai, S. Zhang, Z. Cao et al., “Learning-based prediction of wildfire spread with real-time rate of spread measurement,” Combustion and Flame, vol. 215, pp. 333–341, 2020

    2020

  22. [31]

    An FPGA processor for modelling wildfire spreading,

    P. Progias and G. C. Sirakoulis, “An FPGA processor for modelling wildfire spreading,” Mathematical and Computer Modelling , vol. 57, no. 5, pp. 1436–1452, 2013

    2013

  23. [32]

    Use of spatially refined remote sensing fire detection data to initialize and evaluate coupled weather-wildfire growth model simulations,

    J. L. Coen and W. Schroeder, “Use of spatially refined remote sensing fire detection data to initialize and evaluate coupled weather-wildfire growth model simulations,” Geophys. Res. Lett., vol. 40, pp. 5536–5541, 2013

    2013

  24. [33]

    Improving spatial resolution in exchange of temporal resolution in aliased image sequences,

    L. J. Van Vliet and C. L. L. Hendriks, “Improving spatial resolution in exchange of temporal resolution in aliased image sequences,” in PROCEEDINGS OF THE SCANDINAVIAN CONFERENCE ON IMAGE ANALYSIS, vol. 1, 1999, pp. 493–500

    1999

  25. [34]

    The FLAME dataset: Aerial imagery pile burn detection using drones (uavs),

    A. Shamsoshoara, F. Afghah, A. Razi et al. , “The FLAME dataset: Aerial imagery pile burn detection using drones (uavs),” 2020. [Online]. Available: https://dx.doi.org/10.21227/qad6-r683

    2020 doi

  26. [35]

    Doodlenet: Double deeplab enhanced feature fusion for thermal-color semantic segmenta- tion,

    O. Frigo, L. Martin-Gaff ´e, and C. Wacongne, “Doodlenet: Double deeplab enhanced feature fusion for thermal-color semantic segmenta- tion,” 2022. [Online]. Available: https://arxiv.org/abs/2204.10266

    2022 arXiv

  27. [36]

    Feanet: Feature-enhanced attention network for rgb-thermal real-time semantic segmentation,

    F. Deng, H. Feng, M. Liang, H. Wang, Y . Yang, Y . Gao, J. Chen, J. Hu, X. Guo, and T. L. Lam, “Feanet: Feature-enhanced attention network for rgb-thermal real-time semantic segmentation,” 2021. [Online]. Available: https://arxiv.org/abs/2110.08988

    2021 arXiv

  28. [37]

    FLAME 2: Fire detection and modeling: Aerial multi-spectral image dataset,

    B. Hopkins, L. O’Neill, F. Afghah et al. , “FLAME 2: Fire detection and modeling: Aerial multi-spectral image dataset,” 2022. [Online]. Available: https://dx.doi.org/10.21227/swyw-6j78

    2022 doi

  29. [38]

    IC-GAN: An improved conditional generative adversarial network for RGB-to-IR image translation with applications to forest fire monitoring,

    S. P. Haeri and A. Razi, “IC-GAN: An improved conditional generative adversarial network for RGB-to-IR image translation with applications to forest fire monitoring,” Expert Systems with Applications , vol. 238, p. 121962, 10 2023

    2023

  30. [39]

    Attention-guided synthetic data augmentation for drone-based wildfire detection,

    J. Boone, B. Hopkins, and F. Afghah, “Attention-guided synthetic data augmentation for drone-based wildfire detection,” in IEEE INFOCOM 2023 - IEEE Conference on Computer Communications Workshops (INFOCOM WKSHPS), 2023, pp. 1–6

    2023

  31. [40]

    FLAME diffuser: Grounded wildfire image synthesis using mask guided diffusion,

    H. Wang, S. P. H. Boroujeni, X. Chen et al. , “FLAME diffuser: Grounded wildfire image synthesis using mask guided diffusion,” 2024. [Online]. Available: https://arxiv.org/abs/2403.03463

    2024 arXiv

  32. [41]

    Learning transferable visual models from natural language supervision,

    A. Radford, J. W. Kim, C. Hallacy et al., “Learning transferable visual models from natural language supervision,” 2021. [Online]. Available: https://arxiv.org/abs/2103.00020

    2021 arXiv

  33. [42]

    Thermal image calibration and correction using unpaired cycle-consistent adversarial networks,

    H. Rajoli, P. Afshin, and F. Afghah, “Thermal image calibration and correction using unpaired cycle-consistent adversarial networks,” in2023 57th Asilomar Conference on Signals, Systems, and Computers , 2023, pp. 1425–1429

    2023

  34. [43]

    Unpaired image-to-image translation using cycle-consistent adversarial networks,

    J.-Y . Zhu, T. Park, P. Isola, and A. A. Efros, “Unpaired image-to-image translation using cycle-consistent adversarial networks,” 2020. [Online]. Available: https://arxiv.org/abs/1703.10593

    2020 arXiv

  35. [44]

    Flamefinder: Illuminat- ing obscured fire through smoke with attentive deep metric learning,

    H. Rajoli, S. Khoshdel, F. Afghah, and X. Ma, “Flamefinder: Illuminat- ing obscured fire through smoke with attentive deep metric learning,” IEEE Transactions on Geoscience and Remote Sensing , pp. 1–1, 2024

    2024

  36. [45]

    Hardware acceleration for real-time wildfire detection onboard drone networks,

    A. A. Briley and F. Afghah, “Hardware acceleration for real-time wildfire detection onboard drone networks,” in IEEE INFOCOM 2024 - IEEE Conference on Computer Communications Workshops (INFOCOM WKSHPS), 2024, pp. 01–06

    2024

  37. [46]

    Land and atmosphere precursors to fuel loading, wildfire ignition and post-fire recovery,

    M. R. Alizadeh, J. Adamowski, and D. Entekhabi, “Land and atmosphere precursors to fuel loading, wildfire ignition and post-fire recovery,” Geophysical Research Letters , 01 2024, first published: 19 January

    2024

  38. [47]

    Temporal and spatial monitoring of post-fire forest dynamics using time-series modis data,

    W. Chen, K. Moriya, T. Sakai et al., “Temporal and spatial monitoring of post-fire forest dynamics using time-series modis data,” in 2014 IEEE Geoscience and Remote Sensing Symposium , 2014, pp. 772–775

    2014

  39. [48]

    Multilabel image classification with deep transfer learning for decision support on wildfire response,

    M. Park, D. Q. Tran, S. Lee et al., “Multilabel image classification with deep transfer learning for decision support on wildfire response,” Remote Sensing, vol. 13, no. 19, 2021

    2021

  40. [49]

    Hcp: A flexible cnn framework for multi-label image classification,

    Y . Wei, W. Xia, M. Lin et al. , “Hcp: A flexible cnn framework for multi-label image classification,” IEEE Transactions on Pattern Analysis and Machine Intelligence , vol. 38, no. 9, p. 1901–1907, Sep. 2016. [Online]. Available: http://dx.doi.org/10.1109/TPAMI.2015.2491929

    1901

  41. [50]

    F. Systems. (2021) Free flir thermal dataset for algorithm training. Accessed on: 20 January 2024

    2021

  42. [51]

    Visible-thermal image object detection via the combination of illumination conditions and temperature infor- mation,

    H. Zhou, M. Sun, X. Ren et al., “Visible-thermal image object detection via the combination of illumination conditions and temperature infor- mation,” Remote Sensing, vol. 13, no. 18, 2021

    2021

  43. [52]

    Pixels to pyrometrics: Uas-derived infrared imagery to evaluate and monitor prescribed fire behavior and effects,

    L. O’Neill, P. Fule, A. Watts, C. Moran, B. Hopkins, E. Rowell, A. Thode, and A. Fatemeh, “Pixels to pyrometrics: Uas-derived infrared imagery to evaluate and monitor prescribed fire behavior and effects,” International Journal of Wildland Fire , vol. 33, 2024

    2024

  44. [53]

    R. E. Keane, The photoload sampling technique: estimating surface fuel loadings from downward-looking photographs of synthetic fuelbeds. US Department of Agriculture, Forest Service, Rocky Mountain Research Station, 2007

    2007

  45. [54]

    Metashape professional,

    Agisoft, “Metashape professional,” St. Petersburg, 2023. [Online]. Available: https://www.agisoft.com/

    2023

  46. [55]

    Influence of agisoft metashape parameters on uas structure from motion individual tree detection from canopy height models,

    W. T. Tinkham and N. C. Swayze, “Influence of agisoft metashape parameters on uas structure from motion individual tree detection from canopy height models,” Forests, vol. 12, no. 2, p. 250, 2021

    2021

  47. [56]

    Wildland fire detection and monitoring using a drone-collected RGB/IR image dataset,

    X. Chen, B. Hopkins, H. Wang, L. O’Neill, F. Afghah, A. Razi, P. Ful ´e, J. Coen, E. Rowell, and A. Watts, “Wildland fire detection and monitoring using a drone-collected RGB/IR image dataset,” IEEE Access, vol. 10, pp. 121 301–121 317, 2022

    2022

  48. [2024]

    Available: https://doi.org/10.1029/2023GL105324

    [Online]. Available: https://doi.org/10.1029/2023GL105324

    doi

Pith tools

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