Pith. sign in

REVIEW 6 major objections 5 minor 1 cited by

AI and Generative AI Transforming Disaster Management: A Survey of Damage Assessment and Response Techniques

T0 review · 6 major / 5 minor · reviewed 2026-08-15 · deepseek-v4-flash

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

Pith's one-line read This survey claims to be the first comprehensive review of generative AI techniques for disaster assessment and response.

desk verdict A well-intentioned survey whose citation-to-claim errors in every modality section make it unusable as a reference; the 'first comprehensive' claim also collapses on contact with its own bibliography. read the letter →

arxiv 2505.08202 v1 pith:NAMIAJO6 submitted 2025-05-13 cs.CY cs.AIcs.LG

Aman Raj , Lakshit Arora , Sanjay Surendranath Girija , Shashank Kapoor , Dipen Pradhan , Ankit Shetgaonkar This is my paper · ORCID
classification cs.CYcs.AIcs.LG
keywords GenerativeAIDisasterresponseDamageassessmentMultimodaldataEarthquakeWildfireCycloneMisinformation
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 is a survey of how artificial intelligence, including generative AI, is being applied to damage assessment after natural disasters. Its central claim is that this is the first comprehensive review devoted specifically to GenAI in disaster assessment and response, and that the field now has enough results to map. The authors argue that generative models can combine text, image, video, and audio data to make damage assessment faster and more complete across earthquakes, wildfires, and cyclones. They also catalog the risks, including fake media that could corrupt assessments, privacy concerns, and model bias, and they call for secure and ethical deployment. A sympathetic reader would take this as an organizing reference: a structured inventory of what techniques exist, what evidence supports them, and where research gaps lie.

What carries the argument

The organizing device is a taxonomy that cross-cuts three disaster types — earthquakes, wildfires, and cyclones — with four data modalities: text, image, video, and audio. The technical core is a family of generative models: Variational Autoencoders (VAEs), Generative Adversarial Networks (GANs) such as CycleGAN, Vector-Quantized VAEs (VQ-VAEs), and transformer-based large language models. These models do three main jobs in the survey's account: they generate synthetic training data to compensate for scarce disaster imagery, they translate pre-disaster scenes into simulated post-disaster scenes, and they fuse multimodal inputs for situational awareness. The taxonomy is what carries the survey's argument, because it lets the authors claim coverage of a whole field rather than isolated applications.

What would settle it

Read the abstracts of the cited papers directly and compare them with the survey's in-text claims, starting with [32], [35], and [39]. If a material fraction of the survey's attributions fail this check, its reliability as a map of the field is not established.

Watch

Extended reading notes

Core claim

The paper's central claim, stated in its own words, is that it represents the first comprehensive survey of GenAI techniques used for disaster assessment and response. On the paper's own terms, the discovery is that generative AI has moved from being an abstract possibility to a documented set of applications: VAEs and GANs synthesize training data and simulate damaged scenes, CycleGANs convert pre-fire images to post-fire imagery for wildfire detection, and fine-tuned multimodal models classify building damage from post-earthquake photos. The survey organizes these results by disaster type and by data modality, and it reports guarded evidence, such as GPT-4o achieving 28.6 to 75.0 percent accuracy on EMS-98 damage classification, which it reads as promising but not yet operational. It also argues that the same generative power creates a threat surface, since manipulated images and videos can poison assessment pipelines.

Load-bearing premise

The load-bearing premise is that every citation in the survey actually supports the claim it is attached to, for example that [32] really covers attention mechanisms, [35] really covers video streams, and [39] really covers distress-call analysis.

Editorial extensions

If this is right

  • If the survey's picture is right, damage maps could be produced in near real time by feeding social media text and images, drone video, and audio calls into generative pipelines.
  • Synthetic data from CycleGANs and related models could directly address the chronic shortage of labeled disaster imagery that currently limits supervised damage detection.
  • Fine-tuned multimodal large language models could eventually triage building damage, but the reported accuracy range indicates they are not yet reliable enough for operational use without human review.
  • Because generative media can be weaponized, any deployed assessment system would need provenance checks such as watermarking and perceptual hashing as part of the pipeline.
  • A shared benchmark built from unified multimodal disaster datasets would be needed to measure progress, since the survey identifies the lack of such benchmarks as a gap.

Reading between the lines

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

  • An implication the authors leave implicit is that the survey's practical value depends on its attributions being accurate, so a reader who plans to act on a specific technique should check the original source before relying on the survey's framing.
  • A testable extension of the survey's thesis is to run the same fine-tuned multimodal model used for earthquake damage classification on cyclone and wildfire imagery, to see whether the reported accuracy range generalizes across disaster types.
  • The survey's modality taxonomy suggests a natural benchmark design: fuse text posts, drone video, and audio calls from a single event and measure whether multimodal fusion beats any single modality for damage severity scoring.
  • One consequence the authors gesture at but do not develop is that the same generative models used for assessment could be used for privacy protection, for example by generating sanitized or de-identified versions of sensitive disaster media before analysis.
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

6 major / 5 minor

Summary. This manuscript presents a narrative survey of AI and generative AI applied to disaster damage assessment, with domain-specific reviews for earthquakes, wildfires, and cyclones, and a modality-based organization covering text, image, video, and audio data. It also discusses misinformation, adversarial attacks, privacy, explainability, and future research directions. The paper's stated central contribution is that it is the first comprehensive survey of GenAI techniques for disaster assessment and response. The survey is entirely literature-based; it contains no experiments, datasets, or machine-checked artifacts, so its evidentiary value rests wholly on the accuracy of its citations.

Significance. If its citation base were accurate, this survey would be a useful organizing reference for researchers and practitioners at the intersection of generative AI and disaster management. It correctly identifies a real trend: the use of GANs, VAEs, and large multimodal models for damage classification, synthetic data augmentation, and multimodal fusion. However, the significance is not realized in the current version. The survey's only evidence is its bibliography, and that bibliography is systematically unreliable. Multiple core technical claims across Sections III.B, III.C, III.D, and V are attached to references that do not address the claimed topic. A reader cannot trust the survey's synthesis without independently re-verifying every citation against the primary literature, which defeats the purpose of a survey. For these reasons, the claimed contribution cannot be accepted as it stands.

major comments (6)
  1. [§III.B, refs [32]–[34]] The image-data section cites [32] (Yang 2020, a conditional GAN for aero-engine vibration analysis) to support the claim that attention mechanisms and spatial pyramid pooling extract multi-scale contextual information for damage extraction; [33] (Xing et al., flood vulnerability from remote sensing and street-view imagery) to support image inpainting; and [34] (Ghimire et al., text-based generative AI in construction) to support image super-resolution. None of these references addresses the techniques described in the accompanying sentences. Because image analysis is one of the core modalities of the survey, these mismatches break the evidentiary chain for a central section.
  2. [§III.C, refs [35]–[37]] The video-data section attributes real-time damage classification in video streams to [35] (Dong et al., image super-resolution), building-collapse detection from standalone frames to [36] (Alzantot et al., a synthetic sensor data generator), and SfM/MVS 3D reconstruction to [37] (Viola & Jones, a face detector). These references do not support the specific claims with which they are paired, leaving the entire video-modality discussion without credible evidentiary support.
  3. [§III.D, refs [39]–[41]] The audio-data section attributes distress-call analysis to [39] (Galantucci & Fatiguso, photogrammetry of historical buildings), sound-event detection to [40] (Arbuckle & El Emam, data anonymization), and cross-modal correlation to [41] (Smadi et al., speech recognition). The cited references do not address these claims; in particular, [39] and [40] are topically unrelated to audio-based disaster analysis. The audio section therefore lacks valid literature support for its main technical assertions.
  4. [§V, refs [47]–[48]] The privacy section cites [47] (Zanardelli et al., an image forgery detection survey) as the source for k-anonymity and l-diversity, and [48] (Szegedy et al., intriguing properties of neural networks) as the source for the Laplace mechanism of differential privacy. Neither reference discusses privacy or differential privacy, so the technical discussion of privacy protections is unsupported.
  5. [§III.A, ref [23] vs [29]] In the text-modality section, the manuscript credits 'Tarasconi, Francesco, et al. (2017)' with work on relationship extraction and event correlation, but the citation at that point is [23] (Li et al., Data-Driven Techniques in Disaster Information Management). The actual Tarasconi et al. paper appears as [29]. This is another citation-referent mismatch in a section that otherwise discusses core text-analysis claims.
  6. [Abstract and §VI, novelty claim] The paper's central claim of being 'the first comprehensive survey of GenAI techniques used for disaster assessment and response' is not established. The manuscript itself cites earlier surveys with overlapping scope, notably [4] (Ma et al., generative deep learning in natural hazard analysis) and [9] (Bhadauria, AI/ML in earthquake engineering), but it does not compare its coverage against these works or justify why its scope is distinct. A survey's novelty claim requires a systematic comparison with prior surveys, which is absent.
minor comments (5)
  1. [Reference list, [28] and [30]] References [28] and [30] are identical (both are Havas et al., E2mC). The duplicate [30] is cited in §III.B to support the importance of image data, but the E2mC paper concerns social media and crowdsourcing, not image data.
  2. [Reference list, [25]–[27]] References [25], [26], and [27] appear in the reference list but are never cited in the body of the manuscript.
  3. [§III.A, 'Tarasconi et al. (2017)'] The sentence 'Tarasconi, Francesco, et al(2017) published research suggesting that GenAI also supports relationship extraction' is anachronistic, since GenAI as a term and the underlying large generative models were not the subject of 2017 research; moreover, the cited reference number is wrong, as noted in Major Comment 5.
  4. [Figure 1 caption] Figure 1 is never referenced in the text; the caption 'Taxonomy of Natural Disaster surveys and methods in this survey' does not explain how the figure relates to the surrounding discussion.
  5. [Abstract and title] The abstract contains a subject-verb disagreement: 'AI and Generative AI ... presents a breakthrough solution'; the sentence should read 'present.' Additionally, the reference to 'Gen-AI' with a hyphen in the final sentence of the abstract is inconsistent with the 'GenAI' spelling used elsewhere.

Circularity Check

0 steps flagged · score 0.0 of 10

No circularity: the paper is a narrative survey with no derivation chain, fitted parameters, or load-bearing self-citation.

full rationale

This paper is a narrative survey, not a derivation. It contains no equations, no fitted parameters, no predictions derived from fitted values, and no formal chain of reasoning whose conclusion reduces to its inputs. The central novelty assertion, 'We believe that this work represents the first comprehensive survey of GenAI techniques used for disaster assessment and response,' is a claim about the state of the literature, not a result derived from the paper's own premises or citations. The citation-referent mismatches identified by the reader (e.g., [32] cited for attention mechanisms but referencing aero-engine vibration CGANs, [35] cited for video streams but referencing image super-resolution, [39] cited for distress call analysis but referencing photogrammetry of historical buildings) are serious accuracy and attribution problems that undermine the survey's reliability, but they are not circularity: none of the paper's claims is made true by construction, by renaming a known result, or by citing the authors' own prior work as the sole justification. There is also no fitted-input-called-prediction pattern, because the paper makes no quantitative predictions at all. Under the hard rules, circularity may only be claimed when the paper's own text exhibits a specific reduction (Eq. X = Eq. Y by construction, a fitted parameter renamed as a prediction, or a load-bearing self-citation chain). No such reduction is present. The appropriate finding is therefore no significant circularity, with a score of 0.

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

For a narrative review, there are no free parameters or invented entities. The main hidden assumptions are the accuracy of the references and the representativeness of the selected literature. Both are questionable: citation mismatches are provable, and the selection process is not described.

assumptions (2)
  • domain assumption The cited references accurately support the specific claims they are attached to.
    The survey's credibility depends on correct attribution. This assumption is violated for several references, including [32], [35], [36], and [39], so any conclusion drawn from those citations is unreliable.
  • domain assumption The selected papers are representative of the field of generative AI in disaster damage assessment.
    The survey selects a small, non-systematic set of papers without stating inclusion criteria. Representativeness is assumed but not demonstrated.

how reviews work

0 comments
Cite this review

Pith. "Pith review of AI and Generative AI Transforming Disaster Management: A Survey of Damage Assessment and Response Techniques." pith.science (2026). https://pith.science/paper/NAMIAJO6

@misc{pith2026250508202,
  author       = {Pith},
  title        = {Pith review of: AI and Generative AI Transforming Disaster Management: A Survey of Damage Assessment and Response Techniques},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/NAMIAJO6}},
  note         = {Machine review of arXiv:2505.08202}
}
read the original abstract

Natural disasters, including earthquakes, wildfires and cyclones, bear a huge risk on human lives as well as infrastructure assets. An effective response to disaster depends on the ability to rapidly and efficiently assess the intensity of damage. Artificial Intelligence (AI) and Generative Artificial Intelligence (GenAI) presents a breakthrough solution, capable of combining knowledge from multiple types and sources of data, simulating realistic scenarios of disaster, and identifying emerging trends at a speed previously unimaginable. In this paper, we present a comprehensive review on the prospects of AI and GenAI in damage assessment for various natural disasters, highlighting both its strengths and limitations. We talk about its application to multimodal data such as text, image, video, and audio, and also cover major issues of data privacy, security, and ethical use of the technology during crises. The paper also recognizes the threat of Generative AI misuse, in the form of dissemination of misinformation and for adversarial attacks. Finally, we outline avenues of future research, emphasizing the need for secure, reliable, and ethical Generative AI systems for disaster management in general. We believe that this work represents the first comprehensive survey of Gen-AI techniques being used in the field of Disaster Assessment and Response.

Figures

Figures reproduced from arXiv: 2505.08202 by the authors.

Figure 1
Figure 1. Taxonomy of Natural Disaster surveys and methods in this survey [PITH_FULL_IMAGE:figures/full_fig_p003_1.png] view at source ↗

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. Opportunities and Applications of GenAI in Smart Cities: A User-Centric Survey

    cs.OH 2025-05 conditional novelty 4.0 of 10

    A user-centric survey of generative AI applications for smart cities, covering conversational interfaces for citizens, operators, and planners.

Reference graph

Works this paper leans on

91 extracted references · 36 canonical work pages · cited by 1 Pith paper

  1. [32]

    Conditional Generative Adversarial Networks (CGAN) for Abnormal Vibration of Aero Engine Analysis

    Yang, Lu. “Conditional Generative Adversarial Networks (CGAN) for Abnormal Vibration of Aero Engine Analysis.” 2020 IEEE 2nd International Conference on Civil Aviation Safety and Information Technology (ICCASIT, 2020, pp. 724–28. IEEE Xplore, https://doi.org/10.1109/ICCASIT50869.2020.9368622

    arXiv 2020

  2. [35]

    Image Super-Resolution Using Deep Convolutional Networks

    Dong, Chao, et al. “Image Super-Resolution Using Deep Convolutional Networks.” IEEE Transactions on Pattern Analysis and Machine Intelligence, vol. 38, no. 2, Feb. 2016, pp. 295–307. IEEE Xplore, https://doi.org/10.1109/TPAMI.2015.2439281

    arXiv 2016

  3. [36]

    SenseGen: A Deep Learning Architecture for Synthetic Sensor Data Generation

    Alzantot, Moustafa, et al. “SenseGen: A Deep Learning Architecture for Synthetic Sensor Data Generation.” 2017 IEEE International Conference on Pervasive Computing and Communications Workshops (PerCom Workshops), 2017, pp. 188–93. IEEE Xplore, https://doi.org/10.1109/PERCOMW.2017.7917555

    work page arXiv 2017

  4. [37]

    Viola, Paul & Jones, Michael. (2001). Robust Real-Time Object Detection. International Journal of Computer Vision - IJCV

    work page 2001

  5. [39]

    Advanced Damage Detection Techniques in Historical Buildings Using Digital Photogrammetry and 3D Surface Anlysis

    Galantucci, Rosella Alessia, and Fabio Fatiguso. “Advanced Damage Detection Techniques in Historical Buildings Using Digital Photogrammetry and 3D Surface Anlysis.” Journal of Cultural Heritage, vol. 36, Mar. 2019, pp. 51–62. DOI.org (Crossref), https://doi.org/10.1016/j.culher.2018.09.014

    work page doi:10.1016/j.culher.2018.09.014 2019

  6. [4]

    Generative Deep Learning for Data Generation in Natural Hazard Analysis: Motivations, Advances, Challenges, and Opportunities

    Ma, Zhengjing, et al. “Generative Deep Learning for Data Generation in Natural Hazard Analysis: Motivations, Advances, Challenges, and Opportunities.” Artificial Intelligence Review, vol. 57, no. 6, May 2024, p

    2024

  7. [9]

    Comprehensive Review of AI and ML Tools for Earthquake Damage Assessment and Retrofitting Strategies

    Bhadauria, P. K. S. “Comprehensive Review of AI and ML Tools for Earthquake Damage Assessment and Retrofitting Strategies.” Earth Science Informatics, vol. 17, no. 5, Oct. 2024, pp. 3945–62. DOI.org (Crossref), https://doi.org/10.1007/s12145-024-01431-2

    work page doi:10.1007/s12145-024-01431-2 2024

  8. [31]

    Natural Disasters Detection in Social Media and Satellite Imagery: A Survey

    Said, Naina, et al. “Natural Disasters Detection in Social Media and Satellite Imagery: A Survey.” Multimedia Tools and Applications, vol. 78, no. 22, Nov. 2019, pp. 31267–302. Springer Link, https://doi.org/10.1007/s11042-019-07942-1

    work page doi:10.1007/s11042-019-07942-1 2019

  9. [30]

    E2mC: Improving Emergency Management Service Practice through Social Media and Crowdsourcing Analysis in Near Real Time

    Havas, Clemens, et al. “E2mC: Improving Emergency Management Service Practice through Social Media and Crowdsourcing Analysis in Near Real Time.” Sensors, vol. 17, no. 12, Dec. 2017, p

    work page 2017

  10. [33]

    Springer Link, https://doi.org/10.1186/s41018-016-0013-9

    work page doi:10.1186/s41018-016-0013-9

  11. [34]

    Opportunities and Challenges of Generative AI in Construction Industry: Focusing on Adoption of Text-Based Models

    Ghimire, Prashnna, et al. “Opportunities and Challenges of Generative AI in Construction Industry: Focusing on Adoption of Text-Based Models.” Buildings, vol. 14, no. 1, Jan. 2024, p

    work page 2024

  12. [41]

    Artificial Intelligence for Speech Recognition Based on Neural Networks

    Smadi, Takialddin Al, et al. “Artificial Intelligence for Speech Recognition Based on Neural Networks.” Journal of Signal and Information Processing, vol. 06, no. 02, 2015, pp. 66–72. DOI.org (Crossref), https://doi.org/10.4236/jsip.2015.62006

    work page arXiv 2015

  13. [47]

    Image Forgery Detection: A Survey of Recent Deep-Learning Approaches

    Zanardelli, Marcello, et al. “Image Forgery Detection: A Survey of Recent Deep-Learning Approaches.” Multimedia Tools and Applications, vol. 82, no. 12, May 2023, pp. 17521–66. Springer Link, https://doi.org/10.1007/s11042-022-13797-w

    doi:10.1007/s11042-022-13797-w 2023

  14. [48]

    Intriguing Properties of Neural Networks

    Szegedy, Christian, et al. Intriguing Properties of Neural Networks. arXiv:1312.6199, arXiv, 19 Feb

    arXiv

  15. [23]

    Data-Driven Techniques in Disaster Information Management

    Li, Tao, et al. “Data-Driven Techniques in Disaster Information Management.” ACM Computing Surveys, vol. 50, no. 1, Jan. 2018, pp. 1–45. DOI.org (Crossref), https://doi.org/10.1145/3017678

    work page doi:10.1145/3017678 2018

  16. [29]

    The Role of Unstructured Data in Real-Time Disaster-Related Social Media Monitoring

    Tarasconi, Francesco, et al. “The Role of Unstructured Data in Real-Time Disaster-Related Social Media Monitoring.” 2017 IEEE International Conference on Big Data (Big Data), 2017, pp. 3769–78. IEEE Xplore, https://doi.org/10.1109/BigData.2017.8258377

    arXiv 2017

Show all 91 references
  1. [1]

    Post-earthquake damage assessment using satellite and airborne data in the case of the 1999 Kocaeli earthquake, Turkey

    Ozisik, D., and N. Kerle. "Post-earthquake damage assessment using satellite and airborne data in the case of the 1999 Kocaeli earthquake, Turkey." Proc. of the XXth ISPRS congress: Geo-imagery bridging continents

    1999

  2. [2]

    and Generative Adversarial Networks (GANs) [3], can synthesize representations of disaster-affected areas, enabling rapid analysis and simulation. Generative AI models, particularly task-finetuned ones, also are very well-suited to perform image recognition and classification ...

    2024

  3. [3]

    Generative Adversarial Networks

    Goodfellow, Ian, et al. “Generative Adversarial Networks.” Communications of the ACM, vol. 63, no. 11, Oct. 2020, pp. 139–44. DOI.org (Crossref), https://doi.org/10.1145/3422622

    2020 doi

  4. [5]

    Classifying Earthquake Damage to Buildings Using Machine Learning

    Mangalathu, Sujith, et al. “Classifying Earthquake Damage to Buildings Using Machine Learning.” Earthquake Spectra, vol. 36, no. 1, Feb. 2020, pp. 183–208. DOI.org (Crossref), https://doi.org/10.1177/8755293019878137

    2020 doi

  5. [6]

    UAV-Based Structural Damage Mapping: A Review

    Kerle, Norman, et al. “UAV-Based Structural Damage Mapping: A Review.” ISPRS International Journal of Geo-Information, vol. 9, no. 1, Dec. 2019, p

    2019

  6. [7]

    Gemini and Physical World: Large Language Models Can Estimate the Intensity of Earthquake Shaking from Multimodal Social Media Posts

    Mousavi, S. Mostafa, et al. “Gemini and Physical World: Large Language Models Can Estimate the Intensity of Earthquake Shaking from Multimodal Social Media Posts.” Geophysical Journal International, vol. 240, no. 2, Dec. 2024, pp. 1281–94. DOI.org (Crossref), https://doi.org/1...

    2024 doi

  7. [8]

    Effectiveness of Generative AI for Post-Earthquake Damage Assessment

    Estêvão, João M. C. “Effectiveness of Generative AI for Post-Earthquake Damage Assessment.” Buildings, vol. 14, no. 10, Oct. 2024, p

    2024

  8. [10]

    Remote Sensing Techniques to Assess Active Fire Characteristics and Post-Fire Effects

    Lentile*, Leigh B., et al. “Remote Sensing Techniques to Assess Active Fire Characteristics and Post-Fire Effects.” International Journal of Wildland Fire, vol. 15, no. 3, Sept. 2006, pp. 319–45. www.publish.csiro.au, https://doi.org/10.1071/WF05097

    2006 doi

  9. [11]

    A Review of the Applications of Remote Sensing in Fire Ecology

    Szpakowski, David M., and Jennifer L. R. Jensen. “A Review of the Applications of Remote Sensing in Fire Ecology.” Remote Sensing, vol. 11, no. 22, Jan. 2019, p

    2019

  10. [12]

    A Review on Early Forest Fire Detection Systems Using Optical Remote Sensing

    Barmpoutis, Panagiotis, et al. “A Review on Early Forest Fire Detection Systems Using Optical Remote Sensing.” Sensors, vol. 20, no. 22, Jan. 2020, p

    2020

  11. [14]

    DOI.org (Crossref), https://doi.org/10.3390/ijgi9010014

    doi

  12. [15]

    Wildfire-Detection Method Using DenseNet and CycleGAN Data Augmentation-Based Remote Camera Imagery

    Park, Minsoo, et al. “Wildfire-Detection Method Using DenseNet and CycleGAN Data Augmentation-Based Remote Camera Imagery.” Remote Sensing, vol. 12, no. 22, Nov. 2020, p

    2020

  13. [16]

    Unpaired Image-to-Image Translation Using Cycle-Consistent Adversarial Networks

    Zhu, Jun-Yan, et al. “Unpaired Image-to-Image Translation Using Cycle-Consistent Adversarial Networks.” 2017 IEEE International Conference on Computer Vision (ICCV), 2017, pp. 2242–51. IEEE Xplore, https://doi.org/10.1109/ICCV.2017.244

    2017 doi

  14. [17]

    Cyclone Damage Detection on Building Structures from Pre- and Post-Satellite Images Using Wavelet Based Pattern Recognition

    Radhika, Sudha, et al. “Cyclone Damage Detection on Building Structures from Pre- and Post-Satellite Images Using Wavelet Based Pattern Recognition.” Journal of Wind Engineering and Industrial Aerodynamics, vol. 136, Jan. 2015, pp. 23–33. ScienceDirect, https://doi.org/10.1016...

    2015 doi

  15. [18]

    FloodNet: A High Resolution Aerial Imagery Dataset for Post Flood Scene Understanding

    Rahnemoonfar, Maryam, et al. “FloodNet: A High Resolution Aerial Imagery Dataset for Post Flood Scene Understanding.” IEEE Access, vol. 9, 2021, pp. 89644–54. IEEE Xplore, https://doi.org/10.1109/ACCESS.2021.3090981

    2021

  16. [20]

    Using Artificial Neural Network Models to Assess Hurricane Damage through Transfer Learning

    Calton, Landon, and Zhangping Wei. “Using Artificial Neural Network Models to Assess Hurricane Damage through Transfer Learning.” Applied Sciences, vol. 12, no. 3, Jan. 2022, p

    2022

  17. [21]

    Infrastructure Performance Prediction under Climate-Induced Disasters Using Data Analytics

    Haggag, May, et al. “Infrastructure Performance Prediction under Climate-Induced Disasters Using Data Analytics.” International Journal of Disaster Risk Reduction, vol. 56, Apr. 2021, p. 102121. DOI.org (Crossref), https://doi.org/10.1016/j.ijdrr.2021.102121

    2021

  18. [22]

    Crisis Analytics: Big Data-Driven Crisis Response

    Qadir, Junaid, et al. “Crisis Analytics: Big Data-Driven Crisis Response.” Journal of International Humanitarian Action, vol. 1, no. 1, Aug. 2016, p

    2016

  19. [25]

    Machine Learning Algorithms in Civil Structural Health Monitoring: A Systematic Review

    Flah, Majdi, et al. “Machine Learning Algorithms in Civil Structural Health Monitoring: A Systematic Review.” Archives of Computational Methods in Engineering, vol. 28, no. 4, June 2021, pp. 2621–43. Springer Link, https://doi.org/10.1007/s11831-020-09471-9

    2021 doi

  20. [26]

    Mapping Canopy Damage from Understory Fires in Amazon Forests Using Annual Time Series of Landsat and MODIS Data

    Morton, Douglas C., et al. “Mapping Canopy Damage from Understory Fires in Amazon Forests Using Annual Time Series of Landsat and MODIS Data.” Remote Sensing of Environment, vol. 115, no. 7, July 2011, pp. 1706–20. ScienceDirect, https://doi.org/10.1016/j.rse.2011.03.002

    2011 doi

  21. [27]

    Predicting Forest Fire in the Brazilian Amazon Using MODIS Imagery and Artificial Neural Networks

    Maeda, Eduardo Eiji, et al. “Predicting Forest Fire in the Brazilian Amazon Using MODIS Imagery and Artificial Neural Networks.” International Journal of Applied Earth Observation and Geoinformation, vol. 11, no. 4, Aug. 2009, pp. 265–72. ScienceDirect, https://doi.org/10.1016...

    2009 doi

  22. [38]

    Investigation of Flash Floods on Early Basis: A Factual Comprehensive Review

    Khan, Talha Ahmed, et al. “Investigation of Flash Floods on Early Basis: A Factual Comprehensive Review.” IEEE Access, vol. 8, 2020, pp. 19364–80. IEEE Xplore, https://doi.org/10.1109/ACCESS.2020.2967496

    2020

  23. [42]

    Deepfakes, Misinformation and Disinformation and Authenticity Infrastructure Responses: Impacts on Frontline Witnessing, Distant Witnessing, and Civic Journalism

    Gregory, Sam. “Deepfakes, Misinformation and Disinformation and Authenticity Infrastructure Responses: Impacts on Frontline Witnessing, Distant Witnessing, and Civic Journalism.” Journalism, vol. 23, no. 3, Mar. 2022, pp. 708–29. DOI.org (Crossref), https://doi.org/10.1177/146...

    2022 doi

  24. [43]

    www.mdpi.com, https://doi.org/10.3390/s17122766

    doi

  25. [44]

    A Survey of Detection and Mitigation for Fake Images on Social Media Platforms

    Sharma, Dilip Kumar, et al. “A Survey of Detection and Mitigation for Fake Images on Social Media Platforms.” Applied Sciences, vol. 13, no. 19, Oct. 2023, p. 10980. DOI.org (Crossref), https://doi.org/10.3390/app131910980

    2023 doi

  26. [45]

    Digital Media and Misinformation: An Outlook on Multidisciplinary Strategies against Manipulation

    Caled, Danielle, and Mário J. Silva. “Digital Media and Misinformation: An Outlook on Multidisciplinary Strategies against Manipulation.” Journal of Computational Social Science, vol. 5, no. 1, May 2022, pp. 123–59. Springer Link, https://doi.org/10.1007/s42001-021-00118-8

    2022 doi

  27. [46]

    Flood Vulnerability Assessment of Urban Buildings Based on Integrating High-Resolution Remote Sensing and Street View Images

    Xing, Ziyao, et al. “Flood Vulnerability Assessment of Urban Buildings Based on Integrating High-Resolution Remote Sensing and Street View Images.” Sustainable Cities and Society, vol. 92, May 2023, p. 104467. ScienceDirect, https://doi.org/10.1016/j.scs.2023.104467

    2023

  28. [49]

    Explainable Artificial Intelligence (XAI): What We Know and What Is Left to Attain Trustworthy Artificial Intelligence

    Ali, Sajid, et al. “Explainable Artificial Intelligence (XAI): What We Know and What Is Left to Attain Trustworthy Artificial Intelligence.” Information Fusion, vol. 99, Nov. 2023, p. 101805. DOI.org (Crossref), https://doi.org/10.1016/j.inffus.2023.101805

    2023

  29. [50]

    Multi-Scale Spatial Pyramid Attention Mechanism for Image Recognition: An Effective Approach

    Yu, Yang, et al. “Multi-Scale Spatial Pyramid Attention Mechanism for Image Recognition: An Effective Approach.” Engineering Applications of Artificial Intelligence, vol. 133, July 2024, p. 108261. ScienceDirect, https://doi.org/10.1016/j.engappai.2024.108261

    2024

  30. [51]

    Differential Privacy

    Dwork, Cynthia. “Differential Privacy.” Automata, Languages and Programming, edited by Michele Bugliesi et al., Springer, 2006, pp. 1–12. Springer Link, https://doi.org/10.1007/11787006_1

    2006 doi

  31. [52]

    Interpreting Black-Box Models: A Review on Explainable Artificial Intelligence

    Hassija, Vikas, et al. “Interpreting Black-Box Models: A Review on Explainable Artificial Intelligence.” Cognitive Computation, vol. 16, no. 1, Jan. 2024, pp. 45–74. Springer Link, https://doi.org/10.1007/s12559-023-10179-8

    2024 doi

  32. [54]

    Fairness and Bias in Artificial Intelligence: A Brief Survey of Sources, Impacts, and Mitigation Strategies

    Ferrara, Emilio. “Fairness and Bias in Artificial Intelligence: A Brief Survey of Sources, Impacts, and Mitigation Strategies.” Sci, vol. 6, no. 1, Dec. 2023, p

    2023

  33. [55]

    Improved Wavelet-Based Watermarking through Pixel-Wise Masking

    Barni, M., et al. “Improved Wavelet-Based Watermarking through Pixel-Wise Masking.” IEEE Transactions on Image Processing, vol. 10, no. 5, May 2001, pp. 783–91. DOI.org (Crossref), https://doi.org/10.1109/83.918570

    2001 doi

  34. [56]

    Image Forgery Detection

    Farid, H. “Image Forgery Detection.” IEEE Signal Processing Magazine, vol. 26, no. 2, Mar. 2009, pp. 16–25. DOI.org (Crossref), https://doi.org/10.1109/MSP.2008.931079

    2009

  35. [57]

    models by incorporating a Restricted Boltzmann Machine (RBM) [58], which is optimized with Parallel Tempering in PySA, into the prior encoder step for improving latent space description and segmentation representation [14]. This approach helps to leverage its unique probabilis...

    2020

  36. [58]

    arXiv:1806.07066, arXiv, 19 June

    Restricted Boltzmann Machines: Introduction and Review. arXiv:1806.07066, arXiv, 19 June

    arXiv

  37. [59]

    A Visual Model-Based Perceptual Image Hash for Content Authentication

    Wang, Xiaofeng, et al. “A Visual Model-Based Perceptual Image Hash for Content Authentication.” IEEE Transactions on Information Forensics and Security, vol. 10, no. 7, July 2015, pp. 1336–49. IEEE Xplore, https://doi.org/10.1109/TIFS.2015.2407698

    2015

  38. [60]

    The research confirmed that GAN-generated synthetic data were able to efficiently enhance wildfire detection accuracy

    and ResNet-50 [61]. The research confirmed that GAN-generated synthetic data were able to efficiently enhance wildfire detection accuracy. GenAI's capacity for synthetic training data generation addresses one of the most important issues of wildfire estimation: the unavailabil...

    2025

  39. [62]

    Random Forests

    Breiman, Leo. “Random Forests.” Machine Learning, vol. 45, no. 1, Oct. 2001, pp. 5–32. Springer Link, https://doi.org/10.1023/A:1010933404324

    2001 doi

  40. [68]

    DOI.org (Crossref), https://doi.org/10.3390/sci6010003

    doi

  41. [70]

    Cryptography and network security: principles and practice

    Stallings, William. Cryptography and network security: principles and practice. Seventh edition., Pearson, 2022

    2022

  42. [71]

    U-Net: Convolutional Networks for Biomedical Image Segmentation

    Ronneberger, Olaf, et al. U-Net: Convolutional Networks for Biomedical Image Segmentation. arXiv:1505.04597, arXiv, 18 May

    arXiv

  43. [75]

    (ed.), Musson, R., Schwarz, J., Stucchi, M

    Grünthal, G. (ed.), Musson, R., Schwarz, J., Stucchi, M. (1998): European Macroseismic Scale

    1998

  44. [76]

    DOI.org (Datacite), https://doi.org/10.48550/ARXIV.1409.1556

    doi arXiv

  45. [77]

    DOI.org (Datacite), https://doi.org/10.48550/ARXIV.1512.03385

    doi arXiv

  46. [81]

    DOI.org (Datacite), https://doi.org/10.48550/ARXIV.1704.04861

    doi arXiv

  47. [82]

    DOI.org (Datacite), https://doi.org/10.48550/ARXIV.1905.11946

    1905 doi arXiv

  48. [83]

    DOI.org (Datacite), https://doi.org/10.48550/ARXIV.1707.02921

    doi arXiv

  49. [84]

    DOI.org (Datacite), https://doi.org/10.48550/ARXIV.1807.02758

    doi arXiv

  50. [85]

    DOI.org (Datacite), https://doi.org/10.48550/ARXIV.1506.02640

    doi arXiv

  51. [86]

    Mel Frequency Cepstral Coefficient and Its Applications: A Review

    Abdul, Zrar Kh., and Abdulbasit K. Al-Talabani. “Mel Frequency Cepstral Coefficient and Its Applications: A Review.” IEEE Access, vol. 10, 2022, pp. 122136–58. IEEE Xplore, https://doi.org/10.1109/ACCESS.2022.3223444

    2022

  52. [89]

    DOI.org (Datacite), https://doi.org/10.48550/ARXIV.1705.07874

    doi arXiv

  53. [90]

    DOI.org (Datacite), https://doi.org/10.48550/ARXIV.1602.04938

    doi arXiv

  54. [91]

    Graph Neural Networks: A Review of Methods and Applications

    Zhou, Jie, et al. “Graph Neural Networks: A Review of Methods and Applications.” AI Open, vol. 1, 2020, pp. 57–81. DOI.org (Crossref), https://doi.org/10.1016/j.aiopen.2021.01.001

    2020 doi

  55. [92]

    Geographic Information Systems: Applications and Research Opportunities for Information Systems Researchers

    Mennecke, B. E., and M. D. Crossland. “Geographic Information Systems: Applications and Research Opportunities for Information Systems Researchers.” Proceedings of HICSS-29: 29th Hawaii International Conference on System Sciences, vol. 3, 1996, pp. 537–46 vol.3. IEEE Xplore, h...

    1996

  56. [93]

    DOI.org (Datacite), https://doi.org/10.48550/ARXIV.2112.11739

    doi arXiv

  57. [160]

    Springer Link, https://doi.org/10.1007/s10462-024-10764-9

    doi

  58. [220]

    DOI.org (Crossref), https://doi.org/10.3390/buildings14010220

    doi

  59. [1466]

    DOI.org (Crossref), https://doi.org/10.3390/app12031466

    doi

  60. [2001]

    Greedy Function Approximation: A Gradient Boosting Machine

    “Greedy Function Approximation: A Gradient Boosting Machine.” The Annals of Statistics 29 (5): 1189–1232. https://doi.org/10.1214/aos/1013203451

  61. [2014]

    arXiv.org, https://doi.org/10.48550/arXiv.1312.6199

    doi arXiv

  62. [2015]

    Montufar, Guido

    arXiv.org, https://doi.org/10.48550/arXiv.1505.04597. Montufar, Guido

    doi arXiv

  63. [2016]

    DOI.org (Datacite), https://doi.org/10.48550/ARXIV.1603.02754

    doi arXiv

  64. [2017]

    DOI.org (Datacite), https://doi.org/10.48550/ARXIV.1711.00937

    doi arXiv

  65. [2018]

    arXiv.org, https://doi.org/10.48550/arXiv.1806.07066

    doi arXiv

  66. [2019]

    DOI.org (Datacite), https://doi.org/10.48550/ARXIV.1906.02691

    1906 doi arXiv

  67. [2020]

    DOI.org (Datacite), https://doi.org/10.48550/ARXIV.2010.11929

    2010 doi arXiv

  68. [2021]

    DOI.org (Datacite), https://doi.org/10.48550/ARXIV.2103.14030

    doi arXiv

  69. [2023]

    DOI.org (Datacite), https://doi.org/10.48550/ARXIV.2312.14925

    doi

  70. [2024]

    ntrs.nasa.gov, https://ntrs.nasa.gov/citations/20240008632

  71. [2025]

    Springer Link, https://doi.org/10.1007/s41748-025-00571-9

    doi

  72. [2638]

    www.mdpi.com, https://doi.org/10.3390/rs11222638

    doi

  73. [3255]

    DOI.org (Crossref), https://doi.org/10.3390/buildings14103255

    doi

  74. [3715]

    DOI.org (Crossref), https://doi.org/10.3390/rs12223715

    doi

  75. [6442]

    www.mdpi.com, https://doi.org/10.3390/s20226442

    doi

Pith tools

Reviewed August 15, 2026 · model on record in the stance chip above.