Pith. sign in

REVIEW 6 major objections 5 minor 35 references

A Deep Learning framework for building damage assessment using VHR SAR and geospatial data: demonstration on the 2023 Turkiye Earthquake

T0 review · 6 major / 5 minor · reviewed 2026-08-06 · deepseek-v4-flash

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

Pith's one-line read This paper claims that a single post-event very-high-resolution radar image, combined with building footprints, a digital surface model, and exposure attributes, is enough to detect earthquake-damaged buildings, removing the need for…

desk verdict The new five-city dataset is the real contribution; the cross-city generalization claim is overstated by the paper's own numbers, but it deserves peer review. read the letter →

arxiv 2506.22338 v1 pith:IJYRAOIG submitted 2025-06-27 cs.CV cs.AI

Luigi Russo , Deodato Tapete , Silvia Liberata Ullo , Paolo Gamba This is my paper · ORCID
classification cs.CVcs.AI
keywords buildingdamageassessmentSyntheticApertureRadar(SAR)COSMO-SkyMedmultimodaldeeplearninglatefusionearthquakedetectionfootprintsgeospatialexposuredata
verification ladder T0 review T1 audit T2 compute T3 formal

The pith

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

The reading

The paper claims that building collapse detection after an earthquake does not need pre-disaster imagery or cloud-free optical views: a single post-event very-high-resolution radar image, combined with building footprints, a digital surface model, and exposure attributes, is enough. On five cities hit by the 2023 Türkiye earthquake, the multimodal network reaches an average cross-validation F1 of 0.886 and AUROC of 0.968, and it improves on the radar-plus-footprint baseline in every leave-one-city-out test. If the claim holds, emergency teams could produce building-level damage maps within days of a quake, even under cloud cover and without a pre-event radar archive.

What carries the argument

The carrying mechanism is a modular late-fusion multimodal network: three ResNet-18 encoders independently process the post-event SAR patch, the binary building-footprint mask, and the DSM patch, while a multi-layer perceptron encodes the tabular GEM exposure vector; the four embeddings are concatenated and passed to a two-layer classifier with dropout. The late-fusion design lets the network learn damage signatures from radar while using the auxiliary layers as context, and it allows any modality to be dropped during training or inference without retraining.

What would settle it

Take one of the five cities and re-score the model's per-building predictions against an independent damage inventory, such as field surveys or authoritative post-event optical labeling; if the full multimodal model's F1 against these labels is not better than the SAR-plus-footprint baseline, the claimed gain from geospatial context would be an artifact of the volunteer labels.

Watch

Extended reading notes

Core claim

The central discovery is that structural and socioeconomic context compensates for the absence of pre-event radar data. The authors train a late-fusion network in which a post-event COSMO-SkyMed SAR patch, a building footprint mask, a DSM patch, and a vector of GEM exposure attributes are encoded separately and concatenated for binary damage classification. Adding the DSM raises cross-validation F1 from 0.863 to 0.883, and adding the GEM attributes raises it further to 0.886 with the most stable recall; in unseen-city tests the full configuration improves F1 for all five cities, with the largest gain in Kahramanmaras (0.282 to 0.382). The authors further report that the model flags damage that the volunteer reference labels miss, which they verified with high-resolution optical imagery.

Load-bearing premise

The whole evaluation leans on volunteer-mapped 'destroyed building' polygons as ground truth; if those labels are systematically incomplete or biased, the F1, recall, and generalization numbers are measured against an unreliable standard.

Editorial extensions

If this is right

  • Damage assessment can start from a single post-event SAR acquisition, removing the usual requirement for a pre-disaster radar or optical image of the same area.
  • Adding height and exposure context raises F1 and recall consistently, so the benefit is not confined to one city or one fold.
  • The method generalizes to cities absent from training, which is the operational condition emergency mappers actually face.
  • The automated dataset generation pipeline can be pointed at other disaster regions where footprints and exposure data exist, enabling rapid large-area mapping.
  • The model's tendency to flag damage missed by volunteer annotations suggests it could be used to correct or refine crowdsourced damage maps.

Reading between the lines

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

  • Because the GEM attributes are assigned by nearest centroid from spatially aggregated cells, part of the F1 gain may come from neighborhood-level vulnerability priors rather than per-building radar evidence; ablating GEM or shuffling city labels would test this.
  • If the single-date result transfers to other hazards, the main operational bottleneck shifts from satellite tasking to the availability and quality of building footprints and exposure models, which are often incomplete in the very regions where disasters strike.
  • Reported metrics are computed against volunteer OSM labels that include known false negatives, so the true accuracy may be higher than the headline F1 suggests; an independent validation against field or authoritative post-event surveys is the natural next test.
  • A direct comparison with coherence-based SAR damage proxies, which need pre-event data, on the same five cities would quantify how much accuracy is sacrificed by dropping pre-event acquisitions.
Share X Bluesky LinkedIn Reddit HN

Editorial analysis

A structured set of objections, weighed in public.

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

Referee Report

6 major / 5 minor

Summary. The paper proposes a late-fusion multimodal CNN for binary building damage detection from single-date post-event COSMO-SkyMed VHR SAR patches, augmented with OSM building footprints, CartoSat-1 DSM patches, and GEM exposure attributes. The method is evaluated on a new dataset of 13,487 building-centered patches from five Turkish cities affected by the 2023 Kahramanmaraş earthquake. In 5-fold cross-validation the full configuration achieves F1=0.886 and AUROC=0.968. In leave-one-city-out experiments, F1 ranges from 0.167 to 0.382 depending on the city, and the authors claim that adding DSM and GEM features improves accuracy and generalization. The paper also compares its Islahiye results with the QuickQuakeBuildings benchmark of Sun et al. and finds broadly comparable performance.

Significance. If the claims were fully supported, the framework would be a useful operational contribution: it avoids the need for pre-event VHR SAR or optical imagery, works in cloud-covered conditions, and the data-generation pipeline is automated and extensible. The manuscript contributes a new multi-city dataset and a direct comparison with an existing benchmark, and the in-distribution cross-validation results are internally consistent and show small but consistent improvements from DSM and GEM. The main significance, however, is conditional: the headline generalization numbers are low, the ground-truth labels are acknowledged to be noisy, and the threshold-based metrics are potentially optimistic. These issues must be resolved before the paper's central claims can be accepted at face value.

major comments (6)
  1. [Sec. IV.B and Table III/IV captions] The thresholds are selected by maximizing the F1 score on the precision-recall curve of the test split. This uses the test labels to choose the operating point, which makes the reported precision, recall, F1, and kappa optimistic and complicates comparison with methods using a fixed threshold. Please select thresholds using training folds only, or report metrics at a predefined threshold (e.g., 0.5), and rerun all comparisons.
  2. [Sec. II.B and Fig. 5] The reference data are OSM/HOTOSM destroyed-building polygons, which the paper itself states contain label uncertainty and false negatives; Figure 5 explicitly shows the model identifying damage that OSM labels as intact. All positive-class metrics are therefore computed against an unquantified noisy standard. Please provide a quantitative assessment of label noise, for example by adjudicating a random sample against an independent reference such as Copernicus EMS or high-resolution optical review, and report metrics on the corrected labels. Without this, the claimed improvements from DSM and GEM cannot be cleanly separated from learning the spatial pattern of OSM annotations.
  3. [Sec. IV.D and Table IV] The leave-one-city-out F1 scores are low (0.167–0.382) and the improvements from DSM/GEM are inconsistent across cities. For Islahiye the F1 decreases from 0.284 (SAR+FTP+DSM) to 0.271 (full model), and for Osmaniye recall drops from 0.812 to 0.106 when DSM is added. With only 16 positive examples in Osmaniye and 23 in Türkoğlu, these F1 differences are not statistically meaningful. Please add confidence intervals or significance tests, and temper the abstract/conclusion statement that the model generalizes 'well' to unseen areas; 'partially generalizes with strong city-dependent variation' would be more accurate.
  4. [Table IV] Two rows in Table IV report arithmetically impossible F1 values: the Nurdağı SAR+FTP row gives P=0.240, R=0.240, F1=0.318, and the Osmaniye SAR+FTP+DSM row gives P=0.106, R=0.106, F1=0.143. Since F1 is the harmonic mean of precision and recall, these values cannot be correct as tabulated. Please correct the table or explain the discrepancy.
  5. [Sec. II.C and Table II] The text states that intact buildings were sampled at a 20:1 ratio relative to damaged buildings, but the final dataset contains 12,525 intact versus 962 damaged buildings (approximately 13:1 overall), and Nurdağı has 3,289 intact versus 498 damaged (approximately 6.6:1). Please clarify the exact sampling procedure and report the actual class ratios used in each experiment, because the selected threshold and F1 are sensitive to class prior.
  6. [Sec. IV.E and Table V] The claim that the all-input model 'achieves the best results across key metrics' is not supported by the table: the F1 gain over Sun et al. is 0.013 with overlapping standard deviations, AUROC is lower (0.763 vs. 0.769), and recall is lower. Please rephrase the comparative conclusion as 'comparable' rather than 'better', or provide a formal significance test.
minor comments (5)
  1. [Sec. III and Fig. 4] The dimensionality of the GEM feature vector and the architecture of the MLP (number of layers, hidden units, activations) are not specified; please add these details for reproducibility.
  2. [Sec. IV.B] Cohen's kappa is used in the tables but is never defined in the text; please define it and state how the expected agreement is computed in the imbalanced setting.
  3. [Abstract and Sec. V] The word 'significantly' is used to describe improvements that are not tested for statistical significance; please replace it with a quantitative qualifier such as 'consistently improved' or report significance tests.
  4. [Fig. 5] The caption mentions pre-event imagery, but the framework is intended to use only post-event data; please clarify the source and role of the pre-event panels, since they appear to be for visualization only.
  5. [Sec. IV.A] The statement that all experiments use the same hyperparameter configuration as Sun et al. [27] should be explicit about which hyperparameters (learning rate, batch size, epochs, optimizer, patch size) were adopted.

Circularity Check

1 steps flagged · score 6.0 of 10

Test-set threshold selection makes the headline F1/precision/recall/kappa values partly fitted quantities; AUROC still independently supports the main comparison.

  1. fitted input called prediction [Section IV-B (Performance Metrics); Table IV caption]
    "Classification thresholds were selected based on the value that maximized the F1 score on the precision–recall curve, in order to account for class imbalance. ... Precision, Recall, F1 score, and Cohen's Kappa are reported at the threshold that maximizes the F1 score on the precision–recall curve; AUROC is computed across all thresholds."

    The operating threshold is chosen using the ground-truth labels of the same partition on which F1 is then reported. In the leave-one-city-out experiments this means the F1, precision, recall and kappa for each configuration are evaluated at the threshold that maximizes F1 on the held-out city's own labels; mathematically F1_reported = max_τ F1(τ, y_test, s_model). The reported F1 values and their cross-configuration differences are therefore fit statistics, not out-of-sample predictions at a fixed operating point. The AUROC values (threshold-independent) still provide independent evidence for the model comparison, so the circularity is partial and confined to the threshold-dependent metrics.

full rationale

The derivation chain in this paper is empirical rather than analytic, so most classical circularity patterns do not apply. The model is trained with binary cross-entropy on OSM/HOTOSM-derived labels, and the four input modalities (COSMO-SkyMed SAR patches, OSM footprints, CartoSat DSM patches, and GEM exposure vectors) are genuinely distinct external data sources; no model parameter is fitted to the target metric and then renamed as a prediction. The architecture and baseline follow Sun et al. [27], an external source with no author overlap, so there is no load-bearing self-citation chain. The acknowledged noisiness of the OSM reference labels and the qualitative finding that the model flags damage absent from OSM are label-validity concerns, not circularity: the quantitative target is the OSM label, and whether that label represents true damage is an external-benchmark question, not a reduction of the method to its inputs. The one genuine evaluative reduction is threshold selection: Section IV-B and the Table IV caption state that all threshold-dependent metrics are reported at the threshold maximizing F1 on the precision-recall curve, which in the leave-one-city-out protocol is the held-out city's own ground truth. Reported F1, precision, recall, and kappa are therefore maxima over the test labels rather than predictions at a fixed operating point, making the headline F1 improvements partially constructed. Because AUROC is threshold-independent and improves with the full configuration in all five unseen-city simulations, the central claim retains independent support; the circularity is partial and confined to the threshold-fitted metrics.

Assumptions & free parameters 2 free parameters · 5 assumptions · 0 invented entities

The central claim rests on the reliability of crowdsourced damage labels, on the informativeness of small SAR patches, and on the transferability of neighborhood-scale GEM exposure attributes to individual buildings. The decision threshold is fitted on the test set, and the training set is rebalanced with a 20:1 intact-to-damaged ratio, both of which influence the reported metrics.

free parameters (2)
  • decision threshold = not reported, chosen per test set to maximize F1 on precision-recall curve (Section IV-B)
    The F1, precision, recall, and kappa values in Tables III-V are computed at this threshold, so the reported scores are partly fitted to the test data rather than fixed a priori.
  • sampling ratio (intact:damaged) = 20:1
    All collapsed buildings were kept and intact buildings were randomly sampled at a 1:20 ratio. This design choice affects the training distribution and the resulting precision-recall trade-off, and it varies the effective dataset size across cities, with Osmaniye and Turkoglu contributing very few positive examples.
assumptions (5)
  • domain assumption OSM/HOTOSM 'destroyed buildings' polygons accurately identify collapsed buildings for training and evaluation.
    All labels come from volunteer annotations; the authors acknowledge label uncertainty and manually reviewed ambiguous cases, but the quantitative metrics inherit any systematic errors in these labels. The paper's own Figure 5 shows the model finding damage that OSM labels miss, which means the reference set has false negatives.
  • domain assumption A 32x32 pixel patch (about 80m x 80m at 2.5m resolution) centered on a building contains a detectable SAR signature of collapse.
    The patch was chosen to isolate individual buildings with minimal context; if collapse signatures extend beyond the patch or are masked by layover and shadowing, the model cannot see them.
  • domain assumption Nearest-neighbor assignment of neighborhood-scale GEM attributes to individual buildings preserves the vulnerability signal.
    The GEM dataset provides aggregated statistics per spatial cell; assigning these to each building centroid adds noise, as the authors note, and the model depends on this assignment being informative.
  • domain assumption Building age is a valid proxy for seismic vulnerability.
    The paper explicitly states this assumption (Section II-B) and cites prior literature, but it is a heuristic that can fail for individual buildings.
  • standard math Training a ResNet-18 late fusion model with binary cross-entropy and dropout learns useful representations from the given data.
    No formal guarantees are provided; the method relies on standard empirical machine learning practice.

how reviews work

0 comments
Cite this review

Pith. "Pith review of A Deep Learning framework for building damage assessment using VHR SAR and geospatial data: demonstration on the 2023 Turkiye Earthquake." pith.science (2026). https://pith.science/paper/IJYRAOIG

@misc{pith2026250622338,
  author       = {Pith},
  title        = {Pith review of: A Deep Learning framework for building damage assessment using VHR SAR and geospatial data: demonstration on the 2023 Turkiye Earthquake},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/IJYRAOIG}},
  note         = {Machine review of arXiv:2506.22338}
}
read the original abstract

Building damage identification shortly after a disaster is crucial for guiding emergency response and recovery efforts. Although optical satellite imagery is commonly used for disaster mapping, its effectiveness is often hampered by cloud cover or the absence of pre-event acquisitions. To overcome these challenges, we introduce a novel multimodal deep learning (DL) framework for detecting building damage using single-date very high resolution (VHR) Synthetic Aperture Radar (SAR) imagery from the Italian Space Agency (ASI) COSMO SkyMed (CSK) constellation, complemented by auxiliary geospatial data. Our method integrates SAR image patches, OpenStreetMap (OSM) building footprints, digital surface model (DSM) data, and structural and exposure attributes from the Global Earthquake Model (GEM) to improve detection accuracy and contextual interpretation. Unlike existing approaches that depend on pre and post event imagery, our model utilizes only post event data, facilitating rapid deployment in critical scenarios. The framework effectiveness is demonstrated using a new dataset from the 2023 earthquake in Turkey, covering multiple cities with diverse urban settings. Results highlight that incorporating geospatial features significantly enhances detection performance and generalizability to previously unseen areas. By combining SAR imagery with detailed vulnerability and exposure information, our approach provides reliable and rapid building damage assessments without the dependency from available pre-event data. Moreover, the automated and scalable data generation process ensures the framework's applicability across diverse disaster-affected regions, underscoring its potential to support effective disaster management and recovery efforts. Code and data will be made available upon acceptance of the paper.

Figures

Figures reproduced from arXiv: 2506.22338 by the authors.

Figure 1
Figure 1. Map of the study area showing the location of the Kahramanmaras¸ event supersite in southern T [PITH_FULL_IMAGE:figures/full_fig_p005_1.png] view at source ↗
Figure 2
Figure 2. Summary of GEM exposure metrics for the five analyzed cities. The left panel shows the distribution of structural typologies, such as reinforced [PITH_FULL_IMAGE:figures/full_fig_p006_2.png] view at source ↗
Figure 3
Figure 3. Examples of building-centered image patches from our dataset used [PITH_FULL_IMAGE:figures/full_fig_p006_3.png] view at source ↗
Figures from the paper (3 more)
Figure 4
Figure 4. Figure 4: Schematic of the proposed late-fusion multimodal network for building damage assessment. The architecture processes four input modalities: SAR [PITH_FULL_IMAGE:figures/full_fig_p007_4.png]
Figure 5
Figure 5. Figure 5: Qualitative examples of the model’s performance on cities excluded from training. Samples are drawn from Kahramanmaras¸ (first three rows) [PITH_FULL_IMAGE:figures/full_fig_p009_5.png]
Figure 6
Figure 6. Figure 6: Confusion matrix for the cross-validation fold achieving the highest [PITH_FULL_IMAGE:figures/full_fig_p010_6.png]

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

35 extracted references · 34 canonical work pages

  1. [1]

    Remote Sensing and Earthquake Damage Assessment: Experiences, Limits, and Perspectives,

    F. Dell’Acqua and P. Gamba, “Remote Sensing and Earthquake Damage Assessment: Experiences, Limits, and Perspectives,” Proceedings of the IEEE, vol. 100, no. 10, pp. 2876–2890, 2012

    work page 2012

  2. [2]

    A comprehensive review of earthquake-induced building damage detection with remote sensing techniques,

    L. Dong and J. Shan, “A comprehensive review of earthquake-induced building damage detection with remote sensing techniques,” ISPRS Journal of Photogrammetry and Remote Sensing , vol. 84, pp. 85– 99, 2013. [Online]. Available: https://www.sciencedirect.com/science/ article/pii/S0924271613001627

    work page 2013

  3. [3]

    Earthquake Damage As- sessment of Buildings Using VHR Optical and SAR Imagery,

    D. Brunner, G. Lemoine, and L. Bruzzone, “Earthquake Damage As- sessment of Buildings Using VHR Optical and SAR Imagery,” IEEE Transactions on Geoscience and Remote Sensing , vol. 48, no. 5, pp. 2403–2420, 2010

    work page 2010

  4. [4]

    Assessment of Seismic Building Vulnerability from Space,

    C. Geiß, H. Taubenb ¨ock, S. Tyagunov, A. Tisch, J. Post, and T. Lakes, “Assessment of Seismic Building Vulnerability from Space,”Earthquake Spectra, vol. 30, pp. 1553–1583, 11 2014

    work page 2014

  5. [5]

    A Comparative Study of Texture and Convolutional Neural Network Features for Detecting Collapsed Buildings After Earthquakes Using Pre- and Post-Event Satellite Imagery,

    M. Ji, L. Liu, R. Du, and M. F. Buchroithner, “A Comparative Study of Texture and Convolutional Neural Network Features for Detecting Collapsed Buildings After Earthquakes Using Pre- and Post-Event Satellite Imagery,” Remote Sensing , vol. 11, no. 10, 2019. [Online]. Available: https://www.mdpi.com/2072-4292/11/10/1202

    work page 2019

  6. [6]

    xBD: A Dataset for Assessing Building Damage from Satellite Imagery,

    R. Gupta, R. Hosfelt, S. Sajeev, N. Patel, B. Goodman, J. Doshi, E. Heim, H. Choset, and M. Gaston, “xBD: A Dataset for Assessing Building Damage from Satellite Imagery,” 2019. [Online]. Available: https://arxiv.org/abs/1911.09296

    arXiv 2019

  7. [7]

    Multi-Hazard and Spatial Transferability of a CNN for Automated Building Damage Assessment,

    T. Valentijn, J. Margutti, M. van den Homberg, and J. Laaksonen, “Multi-Hazard and Spatial Transferability of a CNN for Automated Building Damage Assessment,” Remote Sensing , vol. 12, no. 17, 2020. [Online]. Available: https://www.mdpi.com/2072-4292/12/17/2839

    work page 2020

  8. [8]

    Large-scale building damage assessment using a novel hierarchical transformer ar- chitecture on satellite images,

    N. Kaur, C.-C. Lee, A. Mostafavi, and A. Mahdavi-Amiri, “Large-scale building damage assessment using a novel hierarchical transformer ar- chitecture on satellite images,” Computer-Aided Civil and Infrastructure Engineering, vol. 38, 02 2023

    work page 2023

Show all 35 references
  1. [9]

    Automated detection of damaged buildings in post-disaster scenarios: a case study of Kahramanmaras ¸ (T¨urkiye) earthquakes on February 6, 2023,

    C. Serifoglu Yilmaz, V . Yilmaz, K. Tansey, and N. Aljehani, “Automated detection of damaged buildings in post-disaster scenarios: a case study of Kahramanmaras ¸ (T¨urkiye) earthquakes on February 6, 2023,” Natural Hazards, vol. 119, 09 2023

    2023

  2. [10]

    Deep Learning for Building Damage Assessment of the 2023 Turkey Earthquakes,

    J. Malmgren and T. Karlberg, “Deep Learning for Building Damage Assessment of the 2023 Turkey Earthquakes,” KTH Royal Institute of Technology, Master’s thesis, 2023, accessed: 2025- 05-12. [Online]. Available: https://kth.diva-portal.org/smash/get/diva2: 1795168/FULLTEXT01.pdf

    2023

  3. [11]

    Evaluating Deep Learning Based Building Damage Assessment Methods in Densely Built-up Urban Areas: The Case of Kahramanmaras ¸,

    J. Kath, “Evaluating Deep Learning Based Building Damage Assessment Methods in Densely Built-up Urban Areas: The Case of Kahramanmaras ¸,” Master’s thesis, Hertie School, Berlin, 2023, master’s thesis. [Online]. Available: https://github.com/juliankath/Kath Thesis

    2023

  4. [12]

    Deep En- semble Learning for Rapid Large-Scale Postearthquake Damage As- sessment: Application to Satellite Images from the 2023 T ¨urkiye Earth- quakes,

    M. H. Soleimani-Babakamali, M. Askari, M. A. Heravi, R. Sisman, N. Attarchian, A. Askan, R. Soleimani, and E. Taciroglu, “Deep En- semble Learning for Rapid Large-Scale Postearthquake Damage As- sessment: Application to Satellite Images from the 2023 T ¨urkiye Earth- quakes,” ...

    2023

  5. [13]

    Evaluation of Deep Learning Models for Building Damage Mapping in Emergency Response Settings,

    S. Wiguna, B. Adriano, E. Mas, and S. Koshimura, “Evaluation of Deep Learning Models for Building Damage Mapping in Emergency Response Settings,” IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing , vol. 17, pp. 5651–5667, 2024

    2024

  6. [14]

    Evaluating Urban Building Damage of 2023 Kahramanmaras, Turkey Earthquake Sequence Using SAR Change Detection,

    X. Wang, G. Feng, L. He, Q. An, Z. Xiong, H. Lu, W. Wang, N. Li, Y . Zhao, Y . Wang, and Y . Wang, “Evaluating Urban Building Damage of 2023 Kahramanmaras, Turkey Earthquake Sequence Using SAR Change Detection,” Sensors, vol. 23, no. 14, 2023. [Online]. Available: https://www....

    2023

  7. [15]

    The EEFIT Remote Sensing Reconnaissance Mission for the February 2023 Turkey Earthquakes,

    B. V oelker, P. Milillo, A. Tavakkoliestahbanati, V . Macchiarulo, G. Gia- rdina, M. Recla, M. Schmitt, M. Cescon, Y . D. Aktas, and E. So, “The EEFIT Remote Sensing Reconnaissance Mission for the February 2023 Turkey Earthquakes,” IEEE Journal of Selected Topics in Applied Ea...

    2023

  8. [16]

    Deep Learning Meets SAR: Concepts, models, pitfalls, and perspectives,

    X. X. Zhu, S. Montazeri, M. Ali, Y . Hua, Y . Wang, L. Mou, Y . Shi, F. Xu, and R. Bamler, “Deep Learning Meets SAR: Concepts, models, pitfalls, and perspectives,” IEEE Geoscience and Remote Sensing Magazine , vol. 9, no. 4, pp. 143–172, 2021. 12

    2021

  9. [17]

    Change Detection in Heterogeneous Optical and SAR Remote Sensing Images Via Deep Homogeneous Feature Fusion,

    X. Jiang, G. Li, Y . Liu, X.-P. Zhang, and Y . He, “Change Detection in Heterogeneous Optical and SAR Remote Sensing Images Via Deep Homogeneous Feature Fusion,” IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing , vol. 13, pp. 1551– 1566, 2020

    2020

  10. [18]

    CD- TransUNet: A Hybrid Transformer Network for the Change Detection of Urban Buildings Using L-Band SAR Images,

    L. Pang, J. Sun, Y . Chi, Y . Yang, F. Zhang, and L. Zhang, “CD- TransUNet: A Hybrid Transformer Network for the Change Detection of Urban Buildings Using L-Band SAR Images,” Sustainability, vol. 14, no. 16, 2022. [Online]. Available: https://www.mdpi.com/2071-1050/ 14/16/9847

    2022

  11. [19]

    An open-source tool for mapping war destruction at scale in Ukraine using Sentinel-1 time series,

    O. Dietrich, T. Peters, V . Sainte Fare Garnot, V . Sticher, T. Ton- That Whelan, K. Schindler, and J. D. Wegner, “An open-source tool for mapping war destruction at scale in Ukraine using Sentinel-1 time series,” Communications Earth & Environment , vol. 6, no. 1, pp. 1–10, 2025

    2025

  12. [20]

    Detection of Earthquake-Induced Building Damages Using Polarimetric SAR Data,

    S.-E. Park and Y . T. Jung, “Detection of Earthquake-Induced Building Damages Using Polarimetric SAR Data,” Remote Sensing, vol. 12, no. 1,

  13. [21]

    Detection of Damaged Buildings Using Temporal SAR Data with Different Observation Modes,

    M. Kim, S.-E. Park, and S.-J. Lee, “Detection of Damaged Buildings Using Temporal SAR Data with Different Observation Modes,” Remote Sensing , vol. 15, no. 2, 2023. [Online]. Available: https://www.mdpi.com/2072-4292/15/2/308

    2023

  14. [22]

    Optical-to-SAR Translation Based on CDA-GAN for High-Quality Training Sample Generation for Ship Detection in SAR Amplitude Images,

    B. Wu, H. Wang, C. Zhang, and J. Chen, “Optical-to-SAR Translation Based on CDA-GAN for High-Quality Training Sample Generation for Ship Detection in SAR Amplitude Images,” Remote Sensing , vol. 16, no. 16, 2024. [Online]. Available: https://www.mdpi.com/2072-4292/ 16/16/3001

    2024

  15. [23]

    TSGAN: An Optical-to-SAR Dual Conditional GAN for Optical based SAR Temporal Shifting,

    M. Rangzan, S. Attarchi, R. Gloaguen, and S. K. Alavipanah, “TSGAN: An Optical-to-SAR Dual Conditional GAN for Optical based SAR Temporal Shifting,” 2024. [Online]. Available: https: //arxiv.org/abs/2401.00440

    2024 arXiv

  16. [24]

    Unsupervised Domain Adaptation Based on Progressive Transfer for Ship Detection: From Optical to SAR Images,

    Y . Shi, L. Du, Y . Guo, and Y . Du, “Unsupervised Domain Adaptation Based on Progressive Transfer for Ship Detection: From Optical to SAR Images,” IEEE Transactions on Geoscience and Remote Sensing , vol. 60, pp. 1–17, 2022

    2022

  17. [25]

    BRIGHT: A globally distributed multimodal building damage assessment dataset with very-high-resolution for all-weather disaster response,

    H. Chen, J. Song, O. Dietrich, C. Broni-Bediako, W. Xuan, J. Wang, X. Shao, Y . Wei, J. Xia, C. Lan, K. Schindler, and N. Yokoya, “BRIGHT: A globally distributed multimodal building damage assessment dataset with very-high-resolution for all-weather disaster response,” 2025. [...

    2025

  18. [26]

    Global building exposure model for earthquake risk assessment,

    C. Yepes-Estrada, A. Calderon, C. Costa, H. Crowley, J. Dabbeek, M. C. Hoyos, L. Martins, N. Paul, A. Rao, and V . Silva, “Global building exposure model for earthquake risk assessment,” Earthquake Spectra , vol. 39, no. 4, pp. 2212–2235, 2023

    2023

  19. [27]

    QuickQuakeBuildings: Post- Earthquake SAR-Optical Dataset for Quick Damaged-Building Detec- tion,

    Y . Sun, Y . Wang, and M. Eineder, “QuickQuakeBuildings: Post- Earthquake SAR-Optical Dataset for Quick Damaged-Building Detec- tion,” IEEE Geoscience and Remote Sensing Letters , vol. 21, pp. 1–5, 2024

    2024

  20. [28]

    Earthquake building damage detection based on synthetic-aperture-radar imagery and machine learning,

    A. Rao, J. Jung, V . Silva, G. Molinario, and S.-H. Yun, “Earthquake building damage detection based on synthetic-aperture-radar imagery and machine learning,” Natural Hazards and Earth System Sciences , vol. 23, no. 2, pp. 789–807, 2023. [Online]. Available: https: //nhess.co...

    2023

  21. [29]

    COSMO-SkyMed an existing opportunity for observing the Earth,

    F.Covello, F. Battazza, A. Coletta, E. Lopinto, C. Fiorentino, L. Pietranera, G. Valentini, and S. Zoffoli, “COSMO-SkyMed an existing opportunity for observing the Earth,” Journal of Geodynamics , vol. 49, no. 3, pp. 171–180, 2010, wEGENER 2008 - Proceedings of the 14th Genera...

    2010

  22. [30]

    COSMO-SkyMed Mission and Products Description,

    Italian Space Agency (ASI), “COSMO-SkyMed Mission and Products Description,” Italian Space Agency, Rome, Italy, Tech. Rep. ASI-CSM- PMG-NT-001, Issue 2, May 2016, published on 31 May 2016

    2016

  23. [31]

    Macroseismic and mechanical models for the vulnerability and damage assessment of current build- ings,

    S. Lagomarsino and S. Giovinazzi, “Macroseismic and mechanical models for the vulnerability and damage assessment of current build- ings,” Bulletin of Earthquake Engineering , vol. 4, no. 4, pp. 415–443,

  24. [32]

    CG-Net: Conditional GIS-aware Network for Individual Building Segmentation in VHR SAR Images,

    Y . Sun, Y . Hua, L. Mou, and X. Zhu, “CG-Net: Conditional GIS-aware Network for Individual Building Segmentation in VHR SAR Images,” IEEE Transactions on Geoscience and Remote Sensing , 11 2020. Luigi Russo Student Member, IEEE, earned his master’s degree (cum laude) in Elect...

    2020

  25. [1992]

    Member of Academic Senate and PhD Professors’ Board

    Researcher and teacher since 2004 at University of Sannio, Ben- evento (Italy). Member of Academic Senate and PhD Professors’ Board. Courses: Signal theory and elaboration, Telecommunication networks (Bachelor program); Earth monitoring and mission analysis Lab (Master program...

    2004

  26. [2006]

    Available: https://doi.org/10.1007/s10518-006-9024-z

    [Online]. Available: https://doi.org/10.1007/s10518-006-9024-z

    doi

  27. [2020]

    Available: https://www.mdpi.com/2072-4292/12/1/137

    [Online]. Available: https://www.mdpi.com/2072-4292/12/1/137

Pith tools

Reviewed August 6, 2026 · model on record in the stance chip above.