EUVIMEDEUVIMED entdecken
Open Access · CC BY-NC 4.0

FMEformer: a frequency-domain enhanced transformer for cross-domain landslide extraction

Lei Wang, Zhuangzhi Gao, Fang Chen, Nurfashareena Muhamad, Bo Yu

International Journal of Digital Earth · 02.01.2026 · 19 (1)

DOI: 10.1080/17538947.2026.2677973 ↗

Kurz gefasst

Fachartikel in der Zeitschrift International Journal of Digital Earth, online erschienen am 27.05.2026. Verfasst von Lei Wang und 4 weiteren Autor:innen. Die Arbeit nennt 72 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY-NC 4.0). Gefördert von: National Natural Science Foundation of China, Provincial Special Funding for the Construction of Chenzhou National Sustainable Development Agenda Innovation Demonstration Zone, CAS-TWAS Centre of Excellence on Space Technology for Disaster Mitigation.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Landslides are among the most frequent and destructive geological hazards in mountainous regions, posing serious threats to infrastructure and human safety. With the proliferation of high-resolution remote sensing imagery, there is increasing potential for automated landslide extraction. However, most existing methods struggle with limited cross-region generalization and the under-detection of fragmented landslides, due to variations in terrain, scale, and sensor resolution. To overcome these challenges, we propose FMEformer, a frequency-domain–enhanced segmentation framework designed specifically for landslide extraction. The model integrates a Global–Local Frequency Modulation (GLFM) module, which combines Fourier-based global reconstruction with wavelet-based local enhancement guided by local variance, and a Multi-Directional Strip Boundary Enhancement (MD-SBE) module that captures directional boundary features along horizontal, vertical, and diagonal orientations. We evaluated FMEformer against eleven representative deep learning segmentation models, including UNext, UNetFormer, PSPNet, CSUNet, UHRNet, SwinUNet, SegFormer, TransUNet(2021), TransUNet (2024), SKENet, and LSRFormer, covering both CNN-based and Transformer-based architectures. Experiments show that FMEformer improves Intersection over Union (IoU) by at least 6.5%, substantially enhancing landslide extraction accuracy, boundary delineation, and robustness under cross-scenes. This framework provides a scalable solution for precise landslide inventory generation, with strong applicability to disaster risk assessment and emergency response in complex mountainous terrains.

Förderung

  • National Natural Science Foundation of China
  • Provincial Special Funding for the Construction of Chenzhou National Sustainable Development Agenda Innovation Demonstration Zone
  • CAS-TWAS Centre of Excellence on Space Technology for Disaster Mitigation

Zitieren

APA

Wang, L., Gao, Z., Chen, F., Muhamad, N., & Yu, B. (2026). FMEformer: a frequency-domain enhanced transformer for cross-domain landslide extraction. International Journal of Digital Earth, 19(1). https://doi.org/10.1080/17538947.2026.2677973

Vancouver

Wang L, Gao Z, Chen F, Muhamad N, Yu B. FMEformer: a frequency-domain enhanced transformer for cross-domain landslide extraction. International Journal of Digital Earth. 2026;19(1). doi:10.1080/17538947.2026.2677973

BibTeX
@article{Wang2026,
  author = {Wang, Lei and Gao, Zhuangzhi and Chen, Fang and Muhamad, Nurfashareena and Yu, Bo},
  title = {FMEformer: a frequency-domain enhanced transformer for cross-domain landslide extraction},
  journal = {International Journal of Digital Earth},
  year = {2026},
  volume = {19},
  number = {1},
  doi = {10.1080/17538947.2026.2677973}
}
Kostenlos starten →