Pith. sign in

REVIEW 3 major objections 6 minor 67 references

DFPENet-geology: A Deep Learning Framework for High Precision Recognition and Segmentation of Co-seismic Landslides

T0 review · 3 major / 6 minor · reviewed 2026-08-14 · deepseek-v4-flash

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

Pith's one-line read Deep learning scheme recognizes and segments co-seismic landslides from RGB imagery alone, with reported 98.67% pixel consistency in Jiuzhaigou and successful transfer to Hokkaido.

desk verdict Solid Vaihingen segmentation work, but the 98.67% landslide result is internally inconsistent with the described post-processing and likely reflects same-event training overlap. read the letter →

arxiv 1908.10907 v3 pith:P4BNEU2B submitted 2019-08-28 cs.CV

Qingsong Xu , Chaojun Ouyang , Tianhai Jiang , Xuanmei Fan , Duoxiang Cheng This is my paper · ORCID
classification cs.CV
keywords co-seismiclandsliderecognitionsemanticsegmentationDFPENetfeaturepyramidnetworktransferlearningtemporalchangedetectiongeologicfusionremotesensing
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 claims that a single deep segmentation network, DFPENet, combined with landslide shape rules and the difference between pre- and post-earthquake images, can recognize and outline co-seismic landslides from RGB imagery alone with accuracy high enough for emergency response. The authors argue the scheme is cross-scene: a model trained on Chinese landslides, transferred with a small fine-tuning set, also mapped the 2018 Hokkaido landslides. On the ISPRS Vaihingen benchmark the core network reports 93.06% overall accuracy, and consistency with a manually interpreted Jiuzhaigou inventory reaches 98.67% on pixels and 89.95% by landslide count. The stated aim is a rapid, general, RGB-only pipeline for post-earthquake landslide mapping, while explicitly setting aside the landslide boundary error.

What carries the argument

The load-bearing mechanism is a two-stage correction loop: the segmentation network outputs candidate landslide polygons, and the geology module filters them by minimum area (at least four times the spatial resolution) and length-width ratio thresholds, then the pre-event segmentation is subtracted from the post-event segmentation to delete unchanged false positives. The network itself is carried by a dense top-down feature pyramid fed by a ResNet-101 encoder with dilated convolution and atrous spatial pyramid pooling; the feature-filter equations combine gated feature maps so high-level semantic features and low-level detail propagate without redundancy.

What would settle it

A reader could settle this by checking whether any training image, especially the Xiongmaohai-centred Jiuzhaigou samples listed in Table 3, geographically overlaps the 53.6 km² Jiuzhaigou Scenic Area test region; if overlap exists, recompute the consistency after removing those training samples. A second check is to run the same fine-tuning procedure on a held-out earthquake where no landslides from that event appear in training, and compare the resulting mIoU with the reported 98.67%.

Watch

Extended reading notes

Core claim

The paper's central discovery is a pipeline that treats co-seismic landslide recognition as a semantic segmentation task plus a set of cheap, interpretable correction steps. DFPENet is an encoder-decoder with a ResNet-101 dilated backbone, a feature-filter stage using attention gates and gated convolutions, and a dense top-down feature pyramid; it captures small landslides that generic baselines miss. The geology module removes non-landslide regions by minimum-area and length-width ratio thresholds; temporal resolution is added by subtracting the model's pre-earthquake segmentation from its post-earthquake segmentation. The authors report that this raises pixel-level consistency from 92.28% to 98.67% on Jiuzhaigou and, after fine-tuning on 100 patches, transfers to Hokkaido with 77.21% mIoU. Boundary error is explicitly excluded from the headline accuracy.

Load-bearing premise

The headline accuracy assumes the manually interpreted Jiuzhaigou inventory is an independent test, even though the training set deliberately includes landslides from the same 2017 Jiuzhaigou event and an area centered on Xiongmaohai near the test region; if those samples overlap the test scene, the 98.67% number measures scene memorization, not generalization.

Editorial extensions

If this is right

  • If the scheme works as reported, earthquake response teams can produce editable landslide maps from RGB pre/post imagery within minutes of a scene being available.
  • The 93.06% Vaihingen score indicates the same network is a competitive general remote-sensing segmenter, not just a landslide-specific model.
  • Transfer learning from Jiuzhaigou to Hokkaido with 100 patches suggests a trained model can be adapted to new earthquake regions with only small labeled sets, provided at least 20-40% of such a set is available.
  • The temporal subtraction step alone contributes a 6.39% pixel-accuracy gain, implying that unchanged terrain and roads are the main false-positive source in single-date segmentation.

Reading between the lines

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

  • A testable extension is to apply the same geology-plus-temporal pipeline to a multi-event benchmark with held-out earthquake events, measuring per-landslide boundary IoU rather than pixel accuracy; the paper's own caveat that boundary error is not considered suggests boundary-aware metrics would be the first place where the 98.67% number would drop.
  • The length-width and minimum-area filters are prior knowledge encoded as explicit rules; one could train a second network to predict them end-to-end and compare sample efficiency against the hand-tuned thresholds.
  • The cross-scene claim is phrased as RGB-only, but both test sites are mountainous and relatively sparsely vegetated; a stronger test would apply the transferred model to urbanized or densely vegetated earthquake regions, where the temporal subtraction could erroneously remove genuine landslides.
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 / 6 minor

Summary. The manuscript proposes DFPENet, an encoder-decoder semantic segmentation network with a dense feature pyramid module and attention/gated feature filtering, and evaluates it on the ISPRS Vaihingen benchmark, reporting 93.06% overall accuracy. It then introduces DFPENet-geology, a four-phase landslide recognition scheme that combines DFPENet predictions with morphological filtering, temporal subtraction, vectorization, and transfer learning, and applies it to the 2017 Jiuzhaigou earthquake (claimed 98.67% pixel-level consistency) and the 2018 Hokkaido earthquake. The authors claim a new state-of-the-art performance in regional landslide identification while noting that landslide boundary error is not considered. Code and trained models are publicly available on GitHub.

Significance. If the Vaihingen result and the transfer-learning workflow are reproducible, the paper provides a useful segmentation backbone and a practical, RGB-only pipeline for rapid landslide mapping; the component ablation and the public release of models are strengths. However, the central cross-scene generalization claim rests on the Jiuzhaigou consistency analysis, which is compromised by training-set overlap with the test region and by an internally inconsistent fine-tuning row in Table 4. The significance of the landslide claim therefore cannot be assessed without a redesigned evaluation.

major comments (3)
  1. [Table 4 and Section 3.3, Phase 2] Phase 2 states that the temporal fine-tuning step subtracts the pre-event recognition result from the post-event result to remove misidentification areas. Subtraction can only eliminate predicted landslide pixels; it cannot turn previously missed landslide pixels into detected ones. In Table 4, however, the 'Fine tuning results' row increases TP from 915,698 to 924,252 (+8,554) and decreases FN from 20,753 to 12,199 (−8,554), while FP collapses from 55,040 to 245. No described step—morphological filtering, temporal subtraction, or vectorization—adds true positives. This pattern is exactly what one would obtain if an undocumented correction using the validation ground truth was applied, which would make the claimed 98.67% mIoU circular. The authors need to explain how the TP increase is produced, or the result cannot be considered valid.
  2. [Table 3 and Section 4.1.1] The consistency analysis treats the manually interpreted Jiuzhaigou landslide inventory as an external test, but the training set explicitly includes 'Jiuzhaigou seismic landslides around Xiongmaohai' (Table 3), and Section 4.1.1 states that 'the area centered on Xiongmaohai was also added to the training set (Fig. 6)', where Fig. 6 is the geographic location of the research region. If Xiongmaohai lies inside the 53.6 km² Jiuzhaigou Scenic Area test region, the model has seen samples from the test scene during training, and the reported 98.67% is a memorization measure rather than evidence for cross-scene generalization. The validation set also includes Jiuzhaigou seismic landslides, compounding the concern.
  3. [Abstract and Section 4.1.2] The abstract and conclusions claim a new state-of-the-art performance in regional landslide identification while 'not considering the landslide boundary error', yet the comparison to previous methods (Ma et al. 85%, Liu and Wu 97.40%, CDMRF correctness > 0.75) is made without stating whether those methods applied the same boundary treatment. The per-event morphological thresholds are also adapted to each site (minimum area 37.5 m² for Jiuzhaigou and 75 m² for Hokkaido), and no sensitivity analysis is provided for these thresholds or for the length-width ratio threshold. The SOTA claim is therefore not commensurable with the cited baselines as reported.
minor comments (6)
  1. [Section 3.2.2] The abbreviation for Attention Gate Mechanism is introduced as AGM but the text often uses ACM; please make the abbreviations consistent throughout, including in Fig. 1.
  2. [Equations (1)-(3)] The inline formulas are badly garbled by typesetting; the operator symbols and indexing are unclear. Please provide a clean typeset version.
  3. [Section 3.2.5] The dataset description says 'randomly sample the 600×600 patches from the original 33 images' before defining the train/test split; clarify that patches for the test set are sampled only from the 17 test images.
  4. [Section 3.2.5] The pre-trained dataset is described as using '16 images from the training set and 4 images from the validation set', but no validation split has been defined for the 16 training images; please clarify.
  5. [Section 3.2.4 and Section 4.1.2] The GPU is reported as 'Tesla K80' in Section 3.2.4 and as 'Tesla P80' in Sections 3.2.5 and 4.1.2; please correct the inconsistency.
  6. [Section 4.2.2] The text refers to Fig. 10(a)-(f) for the Hokkaido pre/post images and final results, but Fig. 10 is also used earlier for the training-set-size curves; renumber the figures.

Circularity Check

3 steps flagged · score 6.0 of 10

The Jiuzhaigou consistency evaluation is circular because the training set includes the test area and the reported fine-tuning row cannot be produced by the described subtraction step.

  1. fitted input called prediction [Section 4.1.1 (Research regions and data; Table 3)]
    "The research regions were chosen in the Jiuzhaigou Scenic Area, covering an area of 53.6 km2, which is shown in the Fig. 6. ... It is worth noticing that in order to enhance the feature of the Jiuzhaigou landslides, the area centered on Xiongmaohai was also added to the training set (Fig. 6)."

    The paper's cross-scene claim is tested on the Jiuzhaigou Scenic Area, while the same area's Xiongmaohai-centered landslides are explicitly added to the training set. The consistency analysis in Table 4 therefore compares the model against ground truth from the same scene on which it was trained, so the high accuracy reflects memorization of local appearance rather than cross-scene generalization. The reported 98.67% is a scene-specific fitted result, not an external validation.

  2. fitted input called prediction [Section 3.3 Phase 2 and Table 4 (Fine tuning results)]
    "Next, two recognition results are subtracted to remove misidentification areas. ... Fine tuning results 924252 245 12199 99.97% 98.70% 98.67%"

    Pixel-wise subtraction of the pre-event recognition from the post-event recognition can only delete predicted landslide pixels; it cannot create new true positives. The table shows TP rising from 915,698 to 924,252 and FN falling from 20,753 to 12,199 by exactly the same amount, while FP collapses from 55,853 to 245. No described phase (geologic filtering, temporal subtraction, vectorization) adds true positives. The only way to produce this row is to correct predictions with the validation ground truth, making the reported 98.67% mIoU a self-measured result rather than a pipeline output.

1 more flagged steps
  1. fitted input called prediction [Section 4.1.2 (Results of the research region; geologic feature fusion)]
    "Because the study area was dominated by small and medium landslides and the impact of the roads could be eliminated by processing the pre- and post-earthquake images, the feature of the smallest area was then considered as the geological limitation in the geologic feature fusion module (the minimum area was 37.5 m2), thus 573 landslides were obtained."

    The minimum-area threshold and the landslide length-width ratio are chosen after inspecting the study area's own landslides, and the same landslides are then used for the consistency analysis. The gain attributed to the geologic feature fusion module (DFPENet-geology 92.36% vs DFPENet 92.28% in Table 4) is therefore obtained with a post-hoc threshold tuned on the evaluation set; reporting it as a methodological improvement is equivalent to fitting a parameter to the test ground truth.

full rationale

The DFPENet segmentation core has genuinely independent support: the ISPRS Vaihingen benchmark is an external, standard test set, and the reported 93.06% overall accuracy is a self-contained comparison against other published methods. That part is not circular. The circularity is concentrated in the Jiuzhaigou consistency analysis, which carries the paper's central claim of 'state-of-the-art performance in regional landslide identification.' Three concrete reductions are visible in the text: (1) training data include landslides from the same Jiuzhaigou scenic area used as the test region, so the 98.67% consistency score is partly a memorization result; (2) the fine-tuning row in Table 4 is arithmetically inconsistent with the described temporal subtraction, because subtraction cannot increase true positives, implying an undocumented correction using the validation ground truth; and (3) the geologic-feature thresholds are chosen from the same study area and then scored on that area. Because the regional SOTA claim rests on this evaluation, the central claim is substantially forced by the evaluation setup. The Hokkaido experiment is a fine-tuned transfer evaluation rather than zero-shot cross-scene evidence, so it does not repair the circularity. Overall score 6 reflects partial circularity: the network itself has external benchmark support, but the headline regional-landslide result is not an independent prediction.

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

No new physical or conceptual entities are introduced; DFPENet is a neural architecture assembled from cited components. The central empirical claim depends instead on manually chosen thresholds, overlapping training and test inventories, and the assumption that pre-event detections can be subtracted without removing true co-seismic landslides.

free parameters (4)
  • Minimum landslide area threshold (Jiuzhaigou) = 37.5 m2
    Chosen in Phase 1/Section 3.3, Eq. 14 as the lower bound for landslide area; it changes the Jiuzhaigou output from 712 to 573 polygons and is not derived from the network.
  • Minimum landslide area threshold (Hokkaido) = 75 m2
    Set at 75 m2 for the 3 m resolution Planet imagery in Section 4.2.1; this event-specific threshold is applied to the same region that is then presented as the result.
  • Length-width ratio threshold = not specified
    Used in Eq. 14 and Eq. 15 to remove road-like non-landslide regions; the numeric threshold is not given and is tuned per study area.
  • Dilation rates in ResNet-101 encoder = [1, 2, 5]
    Chosen in Section 3.2.1 because the rates were 'validated by experiments'; the Vaihingen results and all downstream landslide results depend on this architectural choice.
assumptions (4)
  • domain assumption The manually interpreted Jiuzhaigou landslide inventory is treated as complete and correct ground truth.
    Section 4.1.2 states ground truth was mainly obtained by artificial visual interpretation and a few investigation reports; all precision/recall numbers in Tables 4 and 5 are measured against it.
  • domain assumption Training on the same earthquake's landslides, including the Xiongmaohai area, does not invalidate the Jiuzhaigou test.
    Section 4.1.1 and Table 3 add Jiuzhaigou seismic landslides around Xiongmaohai to the training set while the research region is the Jiuzhaigou Scenic Area; the paper does not report or discuss the spatial overlap.
  • domain assumption Landslide pixels detected in the pre-earthquake image can all be subtracted as non-co-seismic.
    Phase 2 (Section 3.3) subtracts pre-event recognition results from post-event results; pre-existing landslide scars that reactivated would be removed as false negatives.
  • domain assumption Morphological area and length-width filters capture the geological constraints needed for landslide discrimination.
    Section 3.3 Eq. 14 uses only these two filters despite the text listing NDVI, slope, aspect, curvature and other geological features; this simplification is load-bearing for the claimed false positive reduction.

how reviews work

0 comments
Cite this review

Pith. "Pith review of DFPENet-geology: A Deep Learning Framework for High Precision Recognition and Segmentation of Co-seismic Landslides." pith.science (2026). https://pith.science/paper/P4BNEU2B

@misc{pith2026190810907,
  author       = {Pith},
  title        = {Pith review of: DFPENet-geology: A Deep Learning Framework for High Precision Recognition and Segmentation of Co-seismic Landslides},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/P4BNEU2B}},
  note         = {Machine review of arXiv:1908.10907}
}
read the original abstract

Automatic recognition and segmentation methods now become the essential requirement in identifying co-seismic landslides, which are fundamental for disaster assessment and mitigation in large-scale earthquakes. This approach used to be carried out through pixel-based or object-oriented methods. However, due to the massive amount of remote sensing data, variations in different earthquake scenarios, and the efficiency requirement for post-earthquake rescue, these methods are difficult to develop into an accurate, rapid, comprehensive, and general (cross-scene) solution for co-seismic landslide recognition. This paper develops a robust model, Dense Feature Pyramid with Encoder-decoder Network (DFPENet), to understand and fuse the multi-scale features of objects in remote sensing images. The proposed method achieves a competitive segmentation accuracy on the public ISPRS 2D Semantic. Furthermore, a comprehensive and widely-used scheme is proposed for co-seismic landslide recognition, which integrates image features extracted from the DFPENet model, geologic features, temporal resolution, landslide spatial analysis, and transfer learning, while only RGB images are used. To corroborate its feasibility and applicability, the proposed scheme is applied to two earthquake-triggered landslides in Jiuzhaigou (China) and Hokkaido (Japan), using available pre- and post-earthquake remote sensing images.

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

67 extracted references · 39 canonical work pages

  1. [2]

    IEEE Transactions on Pattern Analysis and Machine Intelligence 39, 2481–2495

    SegNet: A Deep Convolutional Encoder-Decoder Architecture for Image Segmentation. IEEE Transactions on Pattern Analysis and Machine Intelligence 39, 2481–2495. https://doi.org/10.1109/TPAMI.2016.2644615 Barlow, J., Franklin, S.E., Martin, Y.E.,

    arXiv 2016

  2. [6]

    https://doi.org/10.1007/s10346-018-1092-z Yu, F., Koltun, V.,

    Landslides 15, 2521–2524. https://doi.org/10.1007/s10346-018-1092-z Yu, F., Koltun, V.,

    work page doi:10.1007/s10346-018-1092-z

  3. [8]

    arXiv preprint arXiv:1706.05587

    Rethinking Atrous Convolution for Semantic Image Segmentation. arXiv preprint arXiv:1706.05587. Chen, L.C., Zhu, Y., Papandreou, G., Schroff, F., Adam, H., 2018b. Encoder-Decoder with Atrous Separable Convolution for Semantic Image Segmentation, in Proceedings of the European Conference on Computer Vision (ECCV), pp. 801–

    arXiv

  4. [10]

    IEEE Trans

    Automatic Road Detection and Centerline Extraction via Cascaded End-to-End Convolutional Neural Network. IEEE Trans. Geosci. Remote Sensing 55, 3322–3337. https://doi.org/10.1109/TGRS.2017.2669341 Cheng, W., Yang, W., Wang, M., Wang, G., Chen, J.,

    arXiv 2017

  5. [12]

    3014–3017

    Automatic landslide detection from remote sensing images using supervised classification methods, in 2007 IEEE International Geoscience and Remote Sensing Symposium, pp. 3014–3017. https://doi.org/10.1109/IGARSS.2007.4423479 Ding, A., Zhang, Q., Zhou, X., Dai, B.,

    work page arXiv 2007

  6. [13]

    Automatic recognition of landslide based on CNN and texture change detection, in 2016 31st Youth Academic Annual Conference of Chinese Association of Automation (YAC), pp. 444–

    work page 2016

  7. [17]

    https://doi.org/10.13140/2.1.5015.9683 Ghorbanzadeh, O., Blaschke, T., Gholamnia, K., Meena, S.R., Tiede, D., Aryal, J.,

    work page arXiv

  8. [20]

    Deep Residual Learning for Image Recognition, in 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR), pp. 770–778. https://doi.org/10.1109/CVPR.2016.90 Huang, G., Liu, Z., Maaten, L. van der, Weinberger, K.Q.,

    doi:10.1109/cvpr.2016.90 2016

Show all 67 references
  1. [21]

    2261–2269

    Densely Connected Convolutional Networks, in 2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR), pp. 2261–2269. https://doi.org/10.1109/CVPR.2017.243 Huang, R., Zhao, J., Ju, N., Li, G., Lee, M.L., Li, Y.,

    2017 doi

  2. [23]

    Geomorphology 86, 409–440

    Automated classifications of topography from DEMs by an unsupervised nested-means algorithm and a three-part geometric signature. Geomorphology 86, 409–440. https://doi.org/10.1016/j.geomorph.2006.09.012 Japan Meteorological Agency (JMA),

    2006 doi

  3. [28]

    3027–3031

    End-to-end Change Detection Using a Symmetric Fully Convolutional Network for Landslide Mapping, in: ICASSP 2019 - 2019 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), pp. 3027–3031. https://doi.org/10.1109/ICASSP.2019.8682802 Li, H., Xiong, ...

    2019

  4. [29]

    arXiv preprint arXiv:1805.10180

    Pyramid Attention Network for Semantic Segmentation. arXiv preprint arXiv:1805.10180. Li, X., Zhao, H., Han, L., Tong, Y., Yang, K.,

    arXiv

  5. [30]

    arXiv preprint arXiv:1904.01803

    GFF: Gated Fully Fusion for Semantic Segmentation. arXiv preprint arXiv:1904.01803. Lin, D., Shen, D., Shen, S., Ji, Y., Lischinski, D., Cohen-Or, D., Huang, H.,

    1904 arXiv

  6. [31]

    Feature Pyramid Networks for Object Detection, in IEEE Conference on Computer Vision and Pattern Recognition (CVPR), pp. 936–944. https://doi.org/10.1109/CVPR.2017.106 Liu, Y., Fan, B., Wang, L., Bai, J., Xiang, S., Pan, C., 2018a. Semantic Labeling in Very High Resolution Ima...

    2017 doi

  7. [32]

    Procedia Computer Science 91, 566–575

    Geological Disaster Recognition on Optical Remote Sensing Images Using Deep Learning. Procedia Computer Science 91, 566–575. https://doi.org/10.1016/j.procs.2016.07.144 Long, J., Shelhamer, E., Darrell, T.,

    2016 doi

  8. [33]

    Automatic identification of shallow landslides based on Worldview2 remote sensing images. J. Appl. Remote Sens 10, 016008. https://doi.org/10.1117/1.JRS.10.016008 Marmanis, D., Schindler, K., Wegner, J.D., Galliani, S., Datcu, M., Stilla, U.,

    doi

  9. [34]

    ISPRS Journal of Photogrammetry and Remote Sensing 135, 158–172

    Classification With an Edge: Improving Semantic Image Segmentation with Boundary Detection. ISPRS Journal of Photogrammetry and Remote Sensing 135, 158–172. https://doi.org/10.1016/j.isprsjprs.2017.11.009 Martha, T.R., Kerle, N., Jetten, V., van Westen, C.J., Kumar, K.V.,

    2017 doi

  10. [37]

    Remote Sensing of Environment 115, 1743–1757

    Semi-automatic recognition and mapping of rainfall induced shallow landslides using optical satellite images. Remote Sensing of Environment 115, 1743–1757. https://doi.org/10.1016/j.rse.2011.03.006 Mottaghi, R., Chen, X., Liu, X., Cho, N.-G., Lee, S.-W., Fidler, S., Urtasun, R...

    2011 doi

  11. [38]

    The Role of Context for Object Detection and Semantic Segmentation in the Wild, in 2014 IEEE Conference on Computer Vision and Pattern Recognition (CVPR), pp. 891–898. https://doi.org/10.1109/CVPR.2014.119 Nichol, J., Wong, M.S.,

    2014 doi

  12. [40]

    Association for Computational Linguistics, Boulder, Colorado, pp

    Learning Where to Look: Modeling Eye Movements in Reading, in: Proceedings of the Thirteenth Conference on Computational Natural Language Learning (CoNLL-2009). Association for Computational Linguistics, Boulder, Colorado, pp. 93–101. Oktay, O., Schlemper, J., Folgoc, L.L., Le...

    2009

  13. [41]

    arXiv preprint arXiv:1804.03999

    Attention U-Net: Learning Where to Look for the Pancreas. arXiv preprint arXiv:1804.03999. Oquab, M., Bottou, L., Laptev, I., Sivic, J.,

    arXiv

  14. [42]

    1717–1724

    Learning and Transferring Mid-level Image Representations Using Convolutional Neural Networks, in 2014 IEEE Conference on Computer Vision and Pattern Recognition (CVPR), pp. 1717–1724. https://doi.org/10.1109/CVPR.2014.222 Orris, G.J., Williams, J.W.,

    2014 doi

  15. [43]

    Bulletin of the Association of Engineering Geologists 21, 371–375

    Landslide length-width ratios as an aid in landslide identification and verification. Bulletin of the Association of Engineering Geologists 21, 371–375. https://doi.org/10.2113/gseegeosci.xxi.3.371 Paisitkriangkrai, S., Sherrah, J., Janney, P., Hengel, A. van den,

    doi

  16. [44]

    IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing 9, 2868–2881

    Semantic Labeling of Aerial and Satellite Imagery. IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing 9, 2868–2881. https://doi.org/10.1109/JSTARS.2016.2582921 Ronneberger, O., Fischer, P., Brox, T.,

    2016

  17. [47]

    Int J Comput Vis 115, 211–252

    ImageNet Large Scale Visual Recognition Challenge. Int J Comput Vis 115, 211–252. https://doi.org/10.1007/s11263-015-0816-y Sherrah, J.,

    doi

  18. [48]

    arXiv preprint arXiv:1606.02585

    Fully Convolutional Networks for Dense Semantic Labelling of High-Resolution Aerial Imagery. arXiv preprint arXiv:1606.02585. Simonyan, K., Zisserman, A.,

    arXiv

  19. [49]

    arXiv preprint arXiv 1409.1556

    Very Deep Convolutional Networks for Large-Scale Image Recognition. arXiv preprint arXiv 1409.1556. Sun, W., Wang, R.,

    arXiv

  20. [50]

    IEEE Geosci

    Fully Convolutional Networks for Semantic Segmentation of Very High Resolution Remotely Sensed Images Combined With DSM. IEEE Geosci. Remote Sensing Lett. 15, 474–478. https://doi.org/10.1109/LGRS.2018.2795531 Sun, X., Shen, S., Hu, Z.,

    2018

  21. [51]

    Convolutional Neural Networks for Medical Image Analysis: Full Training or Fine Tuning? IEEE Trans. Med. Imaging 35, 1299–1312. https://doi.org/ 10.1109/TMI.2016.2535302 The People’s Government of Sichuan Province,

    2016

  22. [52]

    URL http://www.sc.gov.cn/10462/12771/ 2017/8/14/10430678.shtml (accessed 8.14.19)

    The government report about Jiuzhaigou earthquake. URL http://www.sc.gov.cn/10462/12771/ 2017/8/14/10430678.shtml (accessed 8.14.19). Van Westen, C.J., Castellanos, E., Kuriakose, S.L.,

    2017

  23. [54]

    IEEE Transactions on Geoscience and Remote Sensing 55, 881–893

    Dense Semantic Labeling of Subdecimeter Resolution Images With Convolutional Neural Networks. IEEE Transactions on Geoscience and Remote Sensing 55, 881–893. https://doi.org/10.1109/TGRS.2016.2616585 Wang, F., Fan, X., Yunus, A.P., Siva Subramanian, S., Alonso-Rodriguez, A., D...

    2016

  24. [55]

    Landslides 16, 1551–1566

    Coseismic landslides triggered by the 2018 Hokkaido, Japan (Mw 6.6), earthquake: spatial distribution, controlling factors, and possible failure mechanism. Landslides 16, 1551–1566. https://doi.org/10.1007/s10346-019-01187-7 Wang, H., Wang, Y., Zhang, Q., Xiang, S., Pan, C., 2...

    2018 doi

  25. [57]

    Geoscience Frontiers 6, 825–836

    Preparation of earthquake-triggered landslide inventory maps using remote sensing and GIS technologies: Principles and case studies. Geoscience Frontiers 6, 825–836. https://doi.org/10.1016/j.gsf.2014.03.004 Xu, D., Ouyang, W., Wang, X., Sebe, N.,

    2014 doi

  26. [58]

    PAD-Net: Multi-tasks Guided Prediction-and-Distillation Network for Simultaneous Depth Estimation and Scene Parsing, in 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), pp. 675–684. https://doi.org/10.1109/CVPR.2018.00077 Xu, K., Lei, J., Kiros, R., ...

    2018

  27. [59]

    Show, Attend and Tell: Neural Image CaptionGeneration with Visual Attention, in International Conference on Machine Learning, pp. 2048–

  28. [60]

    Landslides by the 2018 Hokkaido Iburi-Tobu Earthquake on September

    2018

  29. [62]

    https://doi.org/10.1186/s40623-019-1042-6 Krizhevsky, A., Sutskever, I., Hinton, G.E.,

    doi

  30. [63]

    arXiv preprint arXiv:1511.07122

    Multi-scale context aggregation by dilated convolutions. arXiv preprint arXiv:1511.07122

    arXiv

  31. [64]

    Dilated Residual Networks, in IEEE Conference on Computer Vision and Pattern Recognition, pp

    Yu, F., Koltun, V., Funkhouser, T., 2017b. Dilated Residual Networks, in IEEE Conference on Computer Vision and Pattern Recognition, pp. 636–644. https://doi.org/10.1109/CVPR.2017.75 Yu, H., Ma, Y., Wang, L., Zhai, Y., Wang, X.,

    2017 doi

  32. [65]

    A landslide intelligent detection method based on CNN and RSG_R, in 2017 IEEE International Conference on Mechatronics and Automation (ICMA), pp. 40–44. https://doi.org/10.1109/ICMA.2017.8015785 Zeiler, M.D., Fergus, R.,

    2017

  33. [66]

    Visualizing and Understanding Convolutional Networks, in European conference on computer vision, pp. 818–833. https://doi.org/10.1007/978-3-319-10590-1_53 Zeng, X., Ouyang, W., Yan, J., Li, H., Xiao, T., Wang, K., Liu, Y., Zhou, Y., Yang, B., Wang, Z., Zhou, H., Wang, X.,

    doi

  34. [67]

    IEEE Transactions on Pattern Analysis and Machine Intelligence 40, 2109–2123

    Crafting GBD-Net for Object Detection. IEEE Transactions on Pattern Analysis and Machine Intelligence 40, 2109–2123. https://doi.org/10.1109/TPAMI.2017.2745563 Zhang, L., Zhang, L., Du, B.,

    2017

  35. [68]

    IEEE Geoscience and Remote Sensing Magazine 4, 22–40

    Deep Learning for Remote Sensing Data: A Technical Tutorial on the State of the Art. IEEE Geoscience and Remote Sensing Magazine 4, 22–40. https://doi.org/10.1109/MGRS.2016.2540798

    2016

  36. [196]

    https://doi.org/10.3390/rs11020196 Gould, S., Russakovsky, O., Goodfellow, I., Baumstarck, P., Ng, A.Y., Koller, D.,

    doi

  37. [446]

    https://doi.org/10.3390/rs9050446 Xu, C.,

    doi

  38. [448]

    https://doi.org/10.1109/YAC.2016.7804935 Dwivedi, K., Roig, G.,

    2016

  39. [692]

    https://doi.org/10.14358/PERS.72.6.687 Bendale, A., Boult, T.E.,

    doi

  40. [821]

    https://doi.org/10.3390/s18030821 Cheng, G., Wang, Y., Xu, S., Wang, H., Xiang, S., Pan, C.,

    doi

  41. [1158]

    https://doi.org/10.3390/rs11101158 Danneels, G., Pirard, E., Havenith, H.,

    doi

  42. [1850]

    https://doi.org/10.1007/s10346-018-0990-4 Gerke, M.,

    doi

  43. [1984]

    GSA Bulletin 95, 406–421

    Landslides caused by earthquakes. GSA Bulletin 95, 406–421. https://doi.org/10.1130/0016-7606(1984)95<406:LCBE>2.0.CO;2 Kjekstad, O., Highland, L.,

    1984 doi

  44. [1986]

    IEEE Transactions on Pattern Analysis and Machine Intelligence 8, 679–698

    A computational approach to edge detection. IEEE Transactions on Pattern Analysis and Machine Intelligence 8, 679–698. https://doi.org/10.1109/TPAMI.1986.4767851 Chen, L.C., Papandreou, G., Kokkinos, I., Murphy, K., Yuille, A.L., 2018a. DeepLab: Semantic Image Segmentation wit...

    1986

  45. [1988]

    Nature 323, 533–536

    Learning representations by back-propagating errors. Nature 323, 533–536. https://doi.org/10.1038/323533a0 Russakovsky, O., Deng, J., Su, H., Krause, J., Satheesh, S., Ma, S., Huang, Z., Karpathy, A., Khosla, A., Bernstein, M., Berg, A.C., Fei-Fei, L.,

    doi

  46. [2005]

    International Journal of Remote Sensing 26, 1913–1926

    Satellite remote sensing for detailed landslide inventories using change detection and image fusion. International Journal of Remote Sensing 26, 1913–1926. https://doi.org/10.1080/01431160512331314047 Nilsson, M., Nivre, J.,

    1913 doi

  47. [2007]

    International Journal of Remote Sensing 28, 1843–1856

    Comparison between automated and manual mapping of typhoon‐triggered landslides from SPOT‐5 imagery. International Journal of Remote Sensing 28, 1843–1856. https://doi.org/10.1080/01431160600935638 Bruzzone, L., Demir, B.,

    doi

  48. [2008]

    Engineering Geology 102, 112–131

    Spatial data for landslide susceptibility, hazard, and vulnerability assessment: An overview. Engineering Geology 102, 112–131. https://doi.org/10.1016/j.enggeo.2008.03.010 Volpi, M., Tuia, D.,

    2008 doi

  49. [2009]

    Economic and Social Impacts of Landslides, in Landslides – Disaster Risk Reduction, pp. 573–587. https://doi.org/10.1007/978-3-540-69970-5_30 Kobayashi, T., Hayashi, K., Yarai, H.,

    doi

  50. [2010]

    Geomorphology 116, 24–36

    Characterising spectral, spatial and morphometric properties of landslides for semi-automatic detection using object-oriented methods. Geomorphology 116, 24–36. https://doi.org/10.1016/j.geomorph.2009.10.004 Martha, T.R., Kerle, N., van Westen, C.J., Jetten, V., Vinod Kumar, K.,

    2009 doi

  51. [2011]

    Engineering Geology 122, 9–21

    Landslide inventories: The essential part of seismic landslide hazard analyses. Engineering Geology 122, 9–21. https://doi.org/ 10.1016/j.enggeo.2010.06.013 He, K., Zhang, X., Ren, S., Sun, J.,

    2010 doi

  52. [2012]

    ISPRS Journal of Photogrammetry and Remote Sensing 67, 105–119

    Object-oriented analysis of multi-temporal panchromatic images for creation of historical landslide inventories. ISPRS Journal of Photogrammetry and Remote Sensing 67, 105–119. https://doi.org/10.1016/j.isprsjprs.2011.11.004 Minh, N.D., Sang, D.V.,

    2011 doi

  53. [2013]

    Nat Hazards 68, 1021–1039

    Analysis of an anti-dip landslide triggered by the 2008 Wenchuan earthquake in China. Nat Hazards 68, 1021–1039. https://doi.org/10.1007/s11069-013-0671-5 ISPRS, 2016a. 2D Semantic Labeling Challenge. URL http://www2.isprs.org/commissions/comm3/wg4/semantic- labeling.html (acc...

    2008 doi

  54. [2014]

    A Review of Modern Approaches to Classification of Remote Sensing Data, in Land Use and Land Cover Mapping in Europe, pp. 127–143. https://doi.org/10.1007/978-94-007-7969-3_9 Canny, J.,

    doi

  55. [2015]

    (Eds.), Medical Image Computing and Computer-Assisted Intervention – MICCAI 2015, Lecture Notes in Computer Science

    U-Net: Convolutional Networks for Biomedical Image Segmentation, in: Navab, N., Hornegger, J., Wells, W.M., Frangi, A.F. (Eds.), Medical Image Computing and Computer-Assisted Intervention – MICCAI 2015, Lecture Notes in Computer Science. pp. 234–241. https://doi.org/10.1007/97...

    2015 doi

  56. [2016]

    1563–1572

    Towards Open Set Deep Networks, in Proceedings of the IEEE conference on computer vision and pattern recognition (CVPR), pp. 1563–1572. https://doi.org/10.1109/CVPR.2016.173 Bilinski, P., Prisacariu, V., 2018a. Dense Decoder Shortcut Connections for Single-Pass Semantic Segmen...

    2016

  57. [2017]

    Semantic Segmentation of Earth Observation Data Using Multimodal and Multi-scale Deep Networks, in Asian conference on computer vision (ACCV), pp. 180–196. https://doi.org/10.1007/978-3-319-54181-5_12 Badrinarayanan, V., Kendall, A., Cipolla, R.,

    doi

  58. [2018]

    Landslides 15, 967–983

    Coseismic landslides triggered by the 8th August 2017 Ms 7.0 Jiuzhaigou earthquake (Sichuan, China): factors controlling their spatial distribution and implications for the seismogenic blind fault identification. Landslides 15, 967–983. https://doi.org/10.1007/s10346-018-0960-...

    2017 doi

  59. [2019]

    Earth, Planets and Space 71,

    Geodetically estimated location and geometry of the fault plane involved in the 2018 Hokkaido Eastern Iburi earthquake. Earth, Planets and Space 71,

    2018

Pith tools

Reviewed August 14, 2026 · model on record in the stance chip above.