Deep Learning–Based Identification of LULUCF Categories Using a Mixture of Experts Architecture and Remote Sensing Data
DOI:
https://doi.org/10.17398/3101-7177.2.204Palabras clave:
Remote Sensing, Land Use and Land Cover (LULC), LULUCF Monitoring, Deep Learning, Mixture of ExpertsResumen
In the context of climate policies, land use and land cover monitoring is a key component for assessing the LULUCF sector. In this regard, remote sensing combined with deep learning offers new opportunities for more consistent and spatially detailed mapping. The objective of this study is to evaluate the feasibility of a deep learning architecture based on a Mixture of Experts for the hierarchical identification of LULUCF categories, with a particular focus on forested areas. The proposed methodology relies on a cascaded architecture composed of three specialized U-Net models addressing contextual land-use classification, binary tree detection, and tree-species identification, using high-resolution SIOSE-AR data over an MTN50 sheet in northern Spain. Preliminary results indicate a coherent identification of spatial patterns and major land-use categories, although class imbalance remains a limiting factor. Overall, the approach shows strong potential for LULUCF monitoring using remote sensing data.
Descargas
Referencias
Alley, R., Berntsen, T., Bindoff, N.L., Chen, Z., Chidthaisong, A., Friedlingstein, P., Zwiers, F. (2007). Climate change 2007: The physical science basis. Contribution of Working Group I to the Fourth Assessment Report of the Intergovernmental Panel on Climate Change. Summary for Policymakers. IPCC Secretariat, Geneva, Switzerland, 21p.
Eggleston, S., Buendia, L., Miwa, K., Ngara, T., Tanabe, K. (2006). Intergovernmental Panel on Climate Change. 2006 IPCC Guidelines for National Greenhouse Gas Inventories.
Habib, W., Connolly, J. (2023). A national-scale assessment of land use change in peatlands between 1989 and 2020 using Landsat data and Google Earth Engine—a case study of Ireland. Regional Environmental Change, 23(4), 1–16.
Maniatis, D., Dionisio, D., Guarnieri, L., Marchi, G., Mollicone, D., Díaz, A.S.P., Morales, C. (2021). Toward a more representative monitoring of land-use and land-cover dynamics: The use of a sample-based assessment through augmented visual interpretation using open foris collect earth. Remote Sensing, 13(21).
Müller, F., Jaeger, D., Hanewinkel, M. (2019). Digitization in wood supply – A review on how Industry 4.0 will change the forest value chain. Computers and Electronics in Agriculture, 162(April 2018), 206–218.
Radeva, K., Filchev, L., Dimitrov, P., Kirilova, S. (2023). Application of remote sensing to provide support for GHG emissions and removals inventory from LULCF: a case study in Bulgaria. In Ninth International Conference on Remote Sensing and Geoinformation of the Environment (RSCy2023), Vol. 12786, 524–540.
SIOSE, E.T.N. (n.d.). Documento Técnico SIOSE 2011.
United Nations Framework Convention on Climate Change. (2015).
Verburg, P.H., Eickhout, B., Meijl, H. (2008). A multi scale, multi-model approach for analyzing the future dynamics of European land use. Annals of Regional Science, 42(1), 57–77.
Descargas
Publicado
Número
Sección
Licencia
Derechos de autor 2026 Sergio García Martín, Adrián García Bruzón, Fátima Arrogante Funes, Isaac Martín de Diego, Alberto Fernandez Isabel, Ariadna Álvarez Ripado, Fidel Martín González, Víctor Aceña Gil, María del Carmen Lancho Martín, Patricia Arrogante Funes (Autor/a)
Esta obra está bajo una licencia internacional Creative Commons Atribución-NoComercial-CompartirIgual 4.0.
Todo el contenido disponible en el Portal de Revistas-UEx se distribuye bajo una licencia Creative Commons: Atribución 4.0 Internacional (CC BY 4.0)
