Automation of geomorphological mapping has become an important direction in the development of modern geoinformation technologies. Despite recent advances in deep learning, the automated detection of relict periglacial landforms remains challenging because of their subdued morphology and the influence of contemporary land use. This paper proposes a geoinformation framework for automated geomorphological mapping that integrates semantic segmentation, spatial post-processing, and automated generation of vector geospatial data. The framework was evaluated in the Volhynian Upland, a region characterised by diverse relict periglacial landforms. The results indicate that the proposed approach enables automated production of GIS-ready spatial datasets suitable for subsequent expert interpretation and geomorphological mapping.
Barnes, T. J., Schuler, T. V., Filhol, S., & Lilleøren, K. S. (2024). A machine learning approach to the geomorphometric detection of ribbed moraines in Norway. Earth Surface Dynamics, 12, 801-818. https://doi.org/10.5194/esurf-12-801-2024
Bogucki, A., Gołub, B., & Łanczont, M. (2007). Volhynian Upland: The main features of geologic structure and relief. In Materials of the XIV Ukrainian-Polish Seminar 'Problems of the Middle Pleistocene Interglacial' (pp. 6-10). Publishing Centre of Ivan Franko National University of Lviv.
Bogucki, A., Voloshyn, P., & Tomeniuk, O. (2021). Loess cover of the Volhynian Upland: Stratigraphy, key sections, engineering and geological characteristics. Ivan Franko National University of Lviv.
Cao, Z., Tesema, F. B., Fu, P., Ren, J., & Nasr, A. (2026). MCD-Net: A lightweight deep learning baseline for optical-only moraine segmentation. arXiv. https://arxiv.org/abs/2601.02091
Garajeh, M. K., Li, Z., Hasanlu, S., Zare Naghadehi, S., & Haghi, V. H. (2022). Developing an integrated approach based on geographic object-based image analysis and convolutional neural network for volcanic and glacial landforms mapping. Scientific Reports, 12, 21396. https://doi.org/10.1038/s41598-022-26026-z
Maslov, K. A., Persello, C., Schellenberger, T., & Stein, A. (2025). Globally scalable glacier mapping by deep learning matches expert delineation accuracy. Nature Communications, 16, 43. https://doi.org/10.1038/s41467-024-54956-x
Nadachowski, P., Łubniewski, Z., Trzcińska, K., & Tęgowski, J. (2024). Comparison of deep learning approaches in classification of glacial landforms. International Journal of Electronics and Telecommunications, 70(4), 823-829. https://doi.org/10.24425/ijet.2024.152066
Nodjoumi, G., Pozzobon, R., Sauro, F., & Rossi, A. P. (2023). DeepLandforms: A deep learning computer vision toolset applied to a prime use case for mapping planetary skylights. Earth and Space Science, 10, e2022EA002278. https://doi.org/10.1029/2022EA002278
Rocamora, I., Ienco, D., & Ferry, M. (2024). Multi-source deep-learning approach for automatic geomorphological mapping: The case of glacial moraines. Geo-spatial Information Science, 27(6), 1747-1766. https://doi.org/10.1080/10095020.2023.2292587
van der Meij, W. M., Meijles, E. W., Marcos, D., Harkema, T. T. L., Candel, J. H. J., & Maas, G. J. (2022). Comparing geomorphological maps made manually and by deep learning. Earth Surface Processes and Landforms, 47, 1089-1107. https://doi.org/10.1002/esp.5305