CoChat
Sign Up
Cited by 169Landslides and related hazardsRemote-Sensing Image ClassificationRemote Sensing and Land Use

A comprehensive transferability evaluation of U-Net and ResU-Net for landslide detection from Sentinel-2 data (case study areas from Taiwan, China, and Japan)

Open access
Ghorbanzadeh, Omid, Crivellari, Alessandro, Ghamisi, Pedram, Shahabi, Hejar, Blaschke, Thomas·2021·Scientific Reports
DOI 10.1038/s41598-021-94190-9PMC PMC8285525
Open this paper in CoChat — free accountOpen PDF

Abstract

Earthquakes and heavy rainfalls are the two leading causes of landslides around the world. Since they often occur across large areas, landslide detection requires rapid and reliable automatic detection approaches. Currently, deep learning (DL) approaches, especially different convolutional neural network and fully convolutional network (FCN) algorithms, are reliably achieving cutting-edge accuracies in automatic landslide detection. However, these successful applications of various DL approaches have thus far been based on very high resolution satellite images (e.g., GeoEye and WorldView), making it easier to achieve such high detection performances. In this study, we use freely available Sentinel-2 data and ALOS digital elevation model to investigate the application of two well-known FCN algorithms, namely the U-Net and residual U-Net (or so-called ResU-Net), for landslide detection. To our knowledge, this is the first application of FCN for landslide detection only from freely available data. We adapt the algorithms to the specific aim of landslide detection, then train and test with data from three different case study areas located in Western Taitung County (Taiwan), Shuzheng Valley (China), and Eastern Iburi (Japan). We characterize three different window size sample patches to train the algorithms. Our results also contain a comprehensive transferability assessment achieved through different training and testing scenarios in the three case studies. The highest f1-score value of 73.32% was obtained by ResU-Net, trained with a dataset from Japan, and tested on China’s holdout testing area using the sample patch size of 64 × 64 pixels.

The claims and statistics of this paper haven’t been checked yet.

Run the check — free, no account needed. It reads the open-access text and checks the sources it cites.

A few free checks a day · papers already verified open instantly · results are public

Integrity signals

Registry

Crossref · 10.1038/s41598-021-94190-9

Cites

1 of 1 references verified0 citation errors · 0 not checkable

Statistics

Claims the paper makes

The claims and statistics of this paper haven’t been checked yet.

Run the check — free, no account needed. It reads the open-access text and checks the sources it cites.

A few free checks a day · papers already verified open instantly · results are public

Claims other papers make about this one

Each is checked against this paper’s own text: does it say what the citing paper says it does?

  • AI segmentation accuracy was assessed via patch-level cross-validation drawn from the training acquisitions.

    “First, AI segmentation accuracy was assessed via patch-level cross-validation drawn from the training acquisitions; independent validation against field-surveyed vegetation boundaries across distinct climatic and geomorphological zones is required 28 .”

    From AI-driven digital twinning quantifies the trade-off between flood conveyance and riparian stability under climate extremes (2026) · cites this paper as [28]

    The Materials and methods section (Implementation details; page unavailable) says each image was split into separate training and testing areas. The Results section (Scenario 1 and accuracy; page unavailable) says accuracy was assessed using patches from holdout testing areas, not patch-level cross-validation drawn from the training areas. The study concerns landslide detection, not vegetation boundaries. (LLM, conf=0.86)

    Checked by a reasoning model — may be wrong.

    Experimental — accuracy not yet measured

Self-check

Percentages, sums and reported differences re-computed from the paper’s own numbers.

The self-check needs the paper’s full text.

Cites

Flagged first, then by citations

Research works printed but not linked: 56

  1. [1]Guzzetti F Landslide inventory maps: new tools for an old problem Earth Sci. Rev. 2012 112 42 66 10.1016/j.earscirev.2012.02.001 · DOI
  2. [3]Gariano SL Guzzetti F Landslides in a changing climate Earth Sci. Rev. 2016 162 227 252 10.1016/j.earscirev.2016.08.011 · DOI
  3. [4]Sun W Loess landslide inventory map based on GF-1 satellite imagery Remote Sens. Basel 2017 9 314 10.3390/rs9040314 · DOI
  4. [5]Ye C Landslide detection of hyperspectral remote sensing data based on deep learning with constrains IEEE J. Sel. Top. Appl. Earth Obs. Remote Sens. 2019 12 5047 5060 10.1109/JSTARS.2019.2951725 · DOI
  5. [6]Cui P Zhu Y-Y Han Y-S Chen X-Q Zhuang J-Q The 12 May Wenchuan earthquake-induced landslide lakes: distribution and preliminary risk evaluation Landslides 2009 6 209 223 10.1007/s10346-009-0160-9 · DOI
  6. [7]Plank S Twele A Martinis S Landslide mapping in vegetated areas using change detection based on optical and polarimetric sar data Remote Sens. Basel 2016 8 307 10.3390/rs8040307 · DOI
  7. [8]Ghorbanzadeh O Evaluation of different machine learning methods and deep-learning convolutional neural networks for landslide detection Remote Sens. Basel 2019 11 196 10.3390/rs11020196 · DOI
  8. [9]Lima, P. et al. in Workshop on World Landslide Forum. 943–951 (Springer).
  9. [10]Thai Pham B Landslide susceptibility assessment by novel hybrid machine learning algorithms Sustainability 2019 11 4386 10.3390/su11164386 · DOI
  10. [11]Wang H Zhang L Yin K Luo H Li J Landslide identification using machine learning Geosci. Front. 2020 12 351 364 10.1016/j.gsf.2020.02.012 · DOI
  11. [12]Xu, Q., Ouyang, C., Jiang, T., Fan, X. & Cheng, D. DFPENet-geology: a deep learning framework for high precision recognition and segmentation of co-seismic landslides. arXiv preprint https://arxiv.org/abs/1908.10907 (2019). · arXiv
  12. [14]Ghorbanzadeh O An application of Sentinel-1, Sentinel-2, and GNSS data for landslide susceptibility mapping ISPRS Int. J. Geo Inf. 2020 9 561 10.3390/ijgi9100561 · DOI
  13. [15]Siyahghalati S Saraf AK Pradhan B Jebur MN Tehrany MS Rule-based semi-automated approach for the detection of landslides induced by 18 September 2011 Sikkim, Himalaya, earthquake using IRS LISS3 satellite images Geomat. Nat. Haz. Risk 2016 7 326 344 10.1080/19475705.2014.898702 · DOI
  14. [16]Mondini AC Landslide failures detection and mapping using Synthetic Aperture Radar: Past, present and future Earth Sci. Rev. 2021 216 103574 10.1016/j.earscirev.2021.103574 · DOI
  15. [17]Doshida, S. Workshop on World Landslide Forum. 283–287 (Springer).
  16. [18]Ghorbanzadeh O Meena SR Blaschke T Aryal J UAV-Based Slope Failure Detection Using Deep-Learning Convolutional Neural Networks Remote Sens. Basel 2019 11 2046 10.3390/rs11172046 · DOI
  17. [19]Tavakkoli Piralilou S Landslide detection using multi-scale image segmentation and different machine learning models in the higher himalayas Remote Sens. Basel 2019 11 2575 10.3390/rs11212575 · DOI
  18. [21]Shahabi H A semi-automated object-based gully networks detection using different machine learning models: a case study of Bowen catchment, Queensland, Australia Sensors 2019 19 4893 10.3390/s19224893 PMC6891561 31717546 · DOI
  19. [22]Blaschke T Geographic object-based image analysis–towards a new paradigm ISPRS J. Photogramm. Remote. Sens. 2014 87 180 191 10.1016/j.isprsjprs.2013.09.014 24623958 PMC3945831 · DOI
  20. [23]Blaschke T Object based image analysis for remote sensing ISPRS J. Photogramm. Remote. Sens. 2010 65 2 16 10.1016/j.isprsjprs.2009.06.004 PMC3945831 24623958 · DOI
  21. [24]Van Den Eeckhaut M Kerle N Poesen J Hervás J Object-oriented identification of forested landslides with derivatives of single pulse LiDAR data Geomo 2012 173 30 42 10.1016/j.geomorph.2012.05.024 · DOI
  22. [25]Dabiri Z Assessment of landslide-induced geomorphological changes in Hítardalur Valley, Iceland, using Sentinel-1 and Sentinel-2 data Appl. Sci. 2020 10 5848 10.3390/app10175848 · DOI
  23. [26]Bacha AS Van Der Werff H Shafique M Khan H Transferability of object-based image analysis approaches for landslide detection in the Himalaya Mountains of northern Pakistan Int. J. Remote Sens. 2020 41 3390 3410 10.1080/01431161.2019.1701725 · DOI
  24. [27]Chen T Trinder JC Niu R Object-oriented landslide mapping using ZY-3 satellite imagery, random forest and mathematical morphology, for the Three-Gorges Reservoir, China Remote Sens. Basel 2017 9 333 10.3390/rs9040333 · DOI
  25. [28]Jiang X Wang Y Liu W Li S Liu J Capsnet, cnn, fcn: Comparative performance evaluation for image classification Int. J. Mach. Learn. Comput. 2019 9 840 848 10.18178/ijmlc.2019.9.6.881 · DOI
  26. [29]Martinez JAC La Rosa LEC Feitosa RQ Sanches IDA Happ PN Fully convolutional recurrent networks for multidate crop recognition from multitemporal image sequences ISPRS J. Photogramm. Remote Sens. 2021 171 188 201 10.1016/j.isprsjprs.2020.11.007 · DOI
  27. [30]Mboga N Fully convolutional networks and geographic object-based image analysis for the classification of VHR imagery Remote Sens. Basel 2019 11 597 10.3390/rs11050597 · DOI
  28. [31]Chen Z Zhang Y Ouyang C Zhang F Ma J Automated landslides detection for mountain cities using multi-temporal remote sensing imagery Sensors 2018 18 821 10.3390/s18030821 29522424 PMC5876587 · DOI · Paper page
  29. [32]Sameen MI Pradhan B Landslide detection using residual networks and the fusion of spectral and topographic information IEEE Access 2019 7 114363 114373 10.1109/ACCESS.2019.2935761 · DOI
  30. [33]Soares, L. P., Dias, H. C. & Grohmann, C. H. Landslide segmentation with U-Net: evaluating different sampling methods and patch sizes. arXiv preprint https://arxiv.org/abs/2007.06672 (2020). · arXiv
  31. [34]Liu P Wei Y Wang Q Chen Y Xie J Research on post-earthquake landslide extraction algorithm based on improved U-Net model Remote Sens. Basel 2020 12 894 10.3390/rs12050894 · DOI
  32. [35]Qi W Wei M Yang W Xu C Ma C Automatic mapping of landslides by the ResU-Net Remote Sens. Basel 2020 12 2487 10.3390/rs12152487 · DOI
  33. [36]Su Z Deep convolutional neural network–based pixel-wise landslide inventory mapping Landslides 2020 18 1421 1443 10.1007/s10346-020-01557-6 · DOI
  34. [37]Masoud KM Persello C Tolpekin VA Delineation of agricultural field boundaries from sentinel-2 images using a novel super-resolution contour detector based on fully convolutional networks Remote Sens. Basel 2020 12 59 10.3390/rs12010059 · DOI
  35. [38]Osanai N Characteristics of landslides caused by the 2018 Hokkaido Eastern Iburi Earthquake Landslides 2019 16 1517 1528 10.1007/s10346-019-01206-7 · DOI
  36. [39]Kameda J Fluidized landslides triggered by the liquefaction of subsurface volcanic deposits during the 2018 Iburi-Tobu earthquake, Hokkaido Sci. Rep. 2019 9 1 7 10.1038/s41598-019-48820-y 31511623 PMC6739409 · DOI
  37. [40]Yamagishi H Yamazaki F Landslides by the 2018 Hokkaido Iburi-Tobu Earthquake on September 6 Landslides 2018 15 2521 2524 10.1007/s10346-018-1092-z · DOI
  38. [41]Aimaiti Y Liu W Yamazaki F Maruyama Y Earthquake-induced landslide mapping for the 2018 Hokkaido Eastern Iburi earthquake using PALSAR-2 data Remote Sensing 2019 11 2351 10.3390/rs11202351 · DOI
  39. [43]Zhang S Li R Wang F Iio A Characteristics of landslides triggered by the 2018 Hokkaido Eastern Iburi earthquake, Northern Japan Landslides 2019 16 1691 1708 10.1007/s10346-019-01207-6 · DOI
  40. [44]Lin C-W Landslides triggered by the 7 August 2009 Typhoon Morakot in southern Taiwan Eng. Geol. 2011 123 3 12 10.1016/j.enggeo.2011.06.007 · DOI
  41. [45]Lin C-W Recognition of large scale deep-seated landslides in forest areas of Taiwan using high resolution topography J. Asian Earth Sci. 2013 62 389 400 10.1016/j.jseaes.2012.10.022 · DOI
  42. [46]Hu X Hu K Tang J You Y Wu C Assessment of debris-flow potential dangers in the Jiuzhaigou Valley following the August 8, 2017, Jiuzhaigou earthquake, western China Eng. Geol. 2019 256 57 66 10.1016/j.enggeo.2019.05.004 · DOI
  43. [47]Zhao B Landslides and dam damage resulting from the Jiuzhaigou earthquake (8 August 2017), Sichuan, China R. Soc. Open Sci. 2018 5 171418 10.1098/rsos.171418 29657755 PMC5882679 · DOI
  44. [48]Drusch M Sentinel-2: ESA's optical high-resolution mission for GMES operational services Remote Sens. Environ. 2012 120 25 36 10.1016/j.rse.2011.11.026 · DOI
  45. [49]Sentinel, E. User Handbook. ESA Standard Document, Vol. 64.
  46. [50]Long, J., Shelhamer, E. & Darrell, T. in Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition. 3431–3440.
  47. [51]Ronneberger, O., Fischer, P. & Brox, T. in International Conference on Medical Image Computing and Computer-Assisted Intervention. 234–241 (Springer).
  48. [52]Zhang C Identifying and mapping individual plants in a highly diverse high-elevation ecosystem using UAV imagery and deep learning ISPRS J. Photogramm. Remote. Sens. 2020 169 280 291 10.1016/j.isprsjprs.2020.09.025 · DOI
  49. [53]Zhu XX Deep learning in remote sensing: a comprehensive review and list of resources IEEE Geosci. Remote Sens. Mag. 2017 5 8 36 10.1109/MGRS.2017.2762307 · DOI · Paper page
  50. [54]Cui B Fei D Shao G Lu Y Chu J Extracting Raft aquaculture areas from remote sensing images via an improved U-Net with a PSE structure Remote Sens. Basel 2019 11 2053 10.3390/rs11172053 · DOI
  51. [55]Mohammadimanesh F Salehi B Mahdianpari M Gill E Molinier M A new fully convolutional neural network for semantic segmentation of polarimetric SAR imagery in complex land cover ecosystem ISPRS J. Photogramm. Remote. Sens. 2019 151 223 236 10.1016/j.isprsjprs.2019.03.015 · DOI
  52. [56]Abderrahim, N. Y. Q., Abderrahim, S. & Rida, A. in 2020 IEEE International Conference of Moroccan Geomatics (Morgeo). 1–4 (IEEE).
  53. [57]Khryashchev, V. & Larionov, R. in 2020 Moscow Workshop on Electronic and Networking Technologies (MWENT). 1–5 (IEEE).
  54. [59]He, K., Zhang, X., Ren, S. & Sun, J. in Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition. 770–778.
  55. [60]Kingma, D. P. & Ba, J. Adam: A method for stochastic optimization. arXiv preprint https://arxiv.org/abs/1412.6980 (2014). · arXiv
  56. [61]Yi Y Zhang W A new deep-learning-based approach for earthquake-triggered landslide detection from single-temporal RapidEye satellite imagery IEEE J. Sel. Top. Appl. Earth Obs. Remote Sens. 2020 13 6166 6176 10.1109/JSTARS.2020.3028855 · DOI

Cited by

Sorted by citations

Authors
Ghorbanzadeh, O., Crivellari, A., Ghamisi, P., Shahabi, H., & Blaschke, T.
Affiliations
  • University of Salzburg (AT)
  • Helmholtz-Zentrum Dresden-Rossendorf (DE)
  • Slovak Academy of Sciences (SK)
Corresponding author
Ghorbanzadeh, Omid
ORCID
Venue
Scientific Reports
Publisher
Springer Science and Business Media LLC
Published
2021-07-16
Volume
11
Issue
1
Pages
14629-14629
ISSN
2045-2322 (ISSN-L)
DOI
10.1038/s41598-021-94190-9
PMC
PMC8285525
OpenAlex
W3180107394
Cited by
169
References
1
Citations by year
2026: 31 · 2025: 31 · 2024: 33 · 2023: 24 · 2022: 49 · 2021: 1
FWCI
25.14
Type
journal article
Language
en
Open access
Licence
Locations
  • Scientific Reports (publishedVersion, cc-by)
  • PubMed Central (publishedVersion, cc-by)
  • PubMed (publishedVersion)
  • DOAJ (DOAJ: Directory of Open Access Journals) (submittedVersion, cc-by-sa)
Funders
Austrian Science Fund (DK W 1237-N23)
Topics
  • Physical Sciences › Environmental Science › Management, Monitoring, Policy and Law › Landslides and related hazards
  • Physical Sciences › Engineering › Media Technology › Remote-Sensing Image Classification
  • Physical Sciences › Earth and Planetary Sciences › Atmospheric Science › Remote Sensing and Land Use
Keywords
landslide detection, U-Net, ResUNet, fully convolutional network, Sentinel-2, transfer learning, rainfall-induced landslides

About these checks

Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.

Statistics and arithmetic are recomputed, not judged.

Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.

What counts as a claim ›How we measure confidence ›How we check this ›

Are you an author, or do you see a problem with this page? Request a review