Skip to main content

Semantic Segmentation for Landslide Detection Using Segformer

  • Conference paper
  • First Online:
Artificial Intelligence XLI (SGAI 2024)

Abstract

Landslides pose a significant threat to human life and infrastructure which urges the need for efficient techniques for identifying and categorising them. The advent of deep segmentation models such as the Segformer has shown a remarkable empirical performance for semantic segmentation tasks on well-known benchmark datasets, such as ADE20k and Cityscapes. Therefore, this research proposes utilising Segformer on the benchmark Chinese Academy of Sciences (CAS) Landslide Dataset, which features high-quality aerial images of areas impacted or prone to landslides. Taking advantage of the multi-scale attention mechanism and long-range dependency modeling characteristics of the Segformer architecture, this research aims to achieve state-of-the-art results for landslide segmentation using aerial images. Experimental results show the advantage of the Segformer model in segmenting landslide areas, with the largest Segformer variant achieving an Intersection over Union (IoU) score of 87.795% on the Unmanned aerial vehicle (UAV) dataset, surpassing the previous state-of-the-art model, Multiscale Feature Fusion and Enhancement Network (MFFENet), by 3.4%. On the Satellite (SAT) dataset, Segformer attained an IoU score of 79.300%, outperforming the previous best model, DeepLabv3+, by 11.163%. For the combined UAV&SAT dataset, Segformer achieved an IoU score of 85.157%, surpassing DeepLabv3+, the best previous model by 5.032%.

This is a preview of subscription content, log in via an institution to check access.

Access this chapter

Subscribe and save

Springer+ Basic
$34.99 /Month
  • Get 10 units per month
  • Download Article/Chapter or eBook
  • 1 Unit = 1 Article or 1 Chapter
  • Cancel anytime
Subscribe now

Buy Now

Chapter
USD 29.95
Price excludes VAT (USA)
  • Available as PDF
  • Read on any device
  • Instant download
  • Own it forever

Tax calculation will be finalised at checkout

Purchases are for personal use only

Institutional subscriptions

Similar content being viewed by others

References

  1. Bragagnolo, L., et al.: Convolutional neural networks applied to semantic segmentation of landslide scars. CATENA 201 (2021). https://doi.org/10.1016/j.catena.2021.105189

  2. Chen, L.C., et al.: Encoder-decoder with atrous separable convolution for semantic image segmentation. arXiv (2018)

    Google Scholar 

  3. Cheng, L., Li, J., Duan, P., Wang, M.: A small attentional YOLO model for landslide detection from satellite remote sensing images. Landslides 18(8), 2751–2765 (2021). https://doi.org/10.1007/s10346-021-01694-6

    Article  Google Scholar 

  4. Cordts, M., et al.: The cityscapes dataset for semantic urban scene understanding. arXiv (2016)

    Google Scholar 

  5. Dosovitskiy, A., et al.: An image is worth 16x16 words: transformers for image recognition at scale. arXiv (2021)

    Google Scholar 

  6. Froude, M.J., Petley, D.N.: Global fatal landslide occurrence from 2004 to 2016. Nat. Hazards Earth Syst. Sci. 18(8) (2018). https://doi.org/10.5194/nhess-18-2161-2018

  7. Lei, T., et al.: Landslide inventory mapping from bitemporal images using deep convolutional neural networks. IEEE Geosci. Remote Sens. Lett. 16(6) (2019). https://doi.org/10.1109/LGRS.2018.2889307

  8. Long, J., Shelhamer, E., Darrell, T.: Fully convolutional networks for semantic segmentation. arXiv (2015)

    Google Scholar 

  9. Paszke, A., et al.: Pytorch: an imperative style, high-performance deep learning library. arXiv (2019)

    Google Scholar 

  10. Ronneberger, O., Fischer, P., Brox, T.: U-net: Convolutional networks for biomedical image segmentation. arXiv (2015)

    Google Scholar 

  11. Tang, X., et al.: Automatic detection of coseismic landslides using a new transformer method. Remote Sens. 14(12) (2022). https://doi.org/10.3390/rs14122884

  12. Xie, E., et al.: Segformer: Simple and efficient design for semantic segmentation with transformers. arXiv (2021)

    Google Scholar 

  13. Xu, Y., et al.: Cas landslide dataset: a large-scale and multisensor dataset for deep learning-based landslide detection. Sci. Data 11(1) (2024). https://doi.org/10.1038/s41597-023-02847-z

  14. Zhou, B., et al.: Scene parsing through ade20k dataset. In: 2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR). IEEE (2017). https://doi.org/10.1109/CVPR.2017.544

  15. Zhou, W., et al.: Mffenet: multiscale feature fusion and enhancement network for RGB-thermal urban road scene parsing. IEEE Trans. Multimed. 24 (2022). https://doi.org/10.1109/TMM.2021.3086618

Download references

Author information

Authors and Affiliations

Authors

Corresponding author

Correspondence to Hasnain Murtaza Syed .

Editor information

Editors and Affiliations

Rights and permissions

Reprints and permissions

Copyright information

© 2025 The Author(s), under exclusive license to Springer Nature Switzerland AG

About this paper

Check for updates. Verify currency and authenticity via CrossMark

Cite this paper

Syed, H.M., Oghaz, M.M., Saheer, L.B. (2025). Semantic Segmentation for Landslide Detection Using Segformer. In: Bramer, M., Stahl, F. (eds) Artificial Intelligence XLI. SGAI 2024. Lecture Notes in Computer Science(), vol 15447. Springer, Cham. https://doi.org/10.1007/978-3-031-77918-3_3

Download citation

  • DOI: https://doi.org/10.1007/978-3-031-77918-3_3

  • Published:

  • Publisher Name: Springer, Cham

  • Print ISBN: 978-3-031-77917-6

  • Online ISBN: 978-3-031-77918-3

  • eBook Packages: Computer ScienceComputer Science (R0)

Publish with us

Policies and ethics