EUVIMEDEUVIMED entdecken
Open Access · CC BY-NC 4.0

Leveraging vision transformer architectures for glacial and fluvio-glacial landform classification

Paweł Nadachowski, Zbigniew Łubniewski, Karolina Trzcińska, Radosław Wróblewski, Jarosław Tęgowski

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

DOI: 10.1080/17538947.2026.2644685 ↗

Kurz gefasst

Fachartikel in der Zeitschrift International Journal of Digital Earth, online erschienen am 18.03.2026. Verfasst von Paweł Nadachowski und 4 weiteren Autor:innen. Die Arbeit nennt 45 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY-NC 4.0).

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

The study of glacial and fluvio-glacial landforms is essential for understanding the geological history of the Earth. The first step for geomorphologists during their research is the classification of those landforms traditionally performed through manual or semi automatic methods. However, the increasing availability of data in the form of digital elevation models (DEMs) and advancements in neural network architectures may make it possible to automate this process. In this study different Vision Transformer (ViT) architectures were used. The models of this type allowed effective modeling of long-range dependencies in data through the self-attention mechanism, which originated in natural language processing tasks. Their ability to treat grid data as sequences of patches makes them uniquely suited for analyzing the subtle morphological features of glacial landscapes in DEM data. This study investigates the application of ViTs for classifying glacial and fluvio-glacial landforms across various locations and types of terrain. The results demonstrate that ViT architectures achieve a high level of accuracy in this landform classification task.

Zitieren

APA

Nadachowski, P., Łubniewski, Z., Trzcińska, K., Wróblewski, R., & Tęgowski, J. (2026). Leveraging vision transformer architectures for glacial and fluvio-glacial landform classification. International Journal of Digital Earth, 19(1). https://doi.org/10.1080/17538947.2026.2644685

Vancouver

Nadachowski P, Łubniewski Z, Trzcińska K, Wróblewski R, Tęgowski J. Leveraging vision transformer architectures for glacial and fluvio-glacial landform classification. International Journal of Digital Earth. 2026;19(1). doi:10.1080/17538947.2026.2644685

BibTeX
@article{Nadachowski2026,
  author = {Nadachowski, Paweł and Łubniewski, Zbigniew and Trzcińska, Karolina and Wróblewski, Radosław and Tęgowski, Jarosław},
  title = {Leveraging vision transformer architectures for glacial and fluvio-glacial landform classification},
  journal = {International Journal of Digital Earth},
  year = {2026},
  volume = {19},
  number = {1},
  doi = {10.1080/17538947.2026.2644685}
}
Kostenlos starten →