Galileo pretrained remote sensing model – Rwanda crop type classification

The synoptic and temporal capabilities of remote sensing technologies present a powerful solution for crop monitoring, offering continuous and automated assessments throughout various crop growth stages. To leverage this potential, we piloted an automated Artificial Intelligence (AI) remote sensing-...

Descripción completa

Detalles Bibliográficos
Autores principales: Kenduiywo, Benson Kipkemboi, Chemutt, Joseph Kipkoech
Formato: Software
Lenguaje:Inglés
Publicado: 2025
Materias:
Acceso en línea:https://hdl.handle.net/10568/178176
_version_ 1855536753148952576
author Kenduiywo, Benson Kipkemboi
Chemutt, Joseph Kipkoech
author_browse Chemutt, Joseph Kipkoech
Kenduiywo, Benson Kipkemboi
author_facet Kenduiywo, Benson Kipkemboi
Chemutt, Joseph Kipkoech
author_sort Kenduiywo, Benson Kipkemboi
collection Repository of Agricultural Research Outputs (CGSpace)
description The synoptic and temporal capabilities of remote sensing technologies present a powerful solution for crop monitoring, offering continuous and automated assessments throughout various crop growth stages. To leverage this potential, we piloted an automated Artificial Intelligence (AI) remote sensing-based approach by fine-tuning the pipeline for the Galileo Pretrained Remote Sensing Model (https://github.com/nasaharvest/galileo) to map within-season crops in Rwanda. This was accomplished in collaboration with Ministry of Agriculture and Animal Resources (MinAgri) and Rwanda Space Agency (RSA) through a grant support from GIZ. The pilot focused on four Rwandan districts: Nyagatare, Musanze, Nyabihu, and Ruhango, with particular emphasis on consolidated cropland areas (food basket sites) where land parcels are mapped, and crop mapping information is available. This initial pilot mapped four food basket crops namely maize, beans, rice, and Irish potatoes which are part of the Crop Intensification Programme (CIP). By advancing both the development of the foundation model and establishing a clear integration pathway, this project substantially enhanced Rwanda’s agricultural monitoring capabilities, ultimately supporting evidence-based decision-making for food security and climate resilience.
format Software
id CGSpace178176
institution CGIAR Consortium
language Inglés
publishDate 2025
publishDateRange 2025
publishDateSort 2025
record_format dspace
spelling CGSpace1781762025-11-25T13:47:32Z Galileo pretrained remote sensing model – Rwanda crop type classification Kenduiywo, Benson Kipkemboi Chemutt, Joseph Kipkoech artificial intelligence cartography-mapping The synoptic and temporal capabilities of remote sensing technologies present a powerful solution for crop monitoring, offering continuous and automated assessments throughout various crop growth stages. To leverage this potential, we piloted an automated Artificial Intelligence (AI) remote sensing-based approach by fine-tuning the pipeline for the Galileo Pretrained Remote Sensing Model (https://github.com/nasaharvest/galileo) to map within-season crops in Rwanda. This was accomplished in collaboration with Ministry of Agriculture and Animal Resources (MinAgri) and Rwanda Space Agency (RSA) through a grant support from GIZ. The pilot focused on four Rwandan districts: Nyagatare, Musanze, Nyabihu, and Ruhango, with particular emphasis on consolidated cropland areas (food basket sites) where land parcels are mapped, and crop mapping information is available. This initial pilot mapped four food basket crops namely maize, beans, rice, and Irish potatoes which are part of the Crop Intensification Programme (CIP). By advancing both the development of the foundation model and establishing a clear integration pathway, this project substantially enhanced Rwanda’s agricultural monitoring capabilities, ultimately supporting evidence-based decision-making for food security and climate resilience. 2025-11-14 2025-11-25T13:47:32Z 2025-11-25T13:47:32Z Software https://hdl.handle.net/10568/178176 en Open Access Kenduiywo, B.K.; Chemutt, J.K. (2025) Galileo pretrained remote sensing model – Rwanda crop type classification. [Software] Github. Published online 11 November 2025.
spellingShingle artificial intelligence
cartography-mapping
Kenduiywo, Benson Kipkemboi
Chemutt, Joseph Kipkoech
Galileo pretrained remote sensing model – Rwanda crop type classification
title Galileo pretrained remote sensing model – Rwanda crop type classification
title_full Galileo pretrained remote sensing model – Rwanda crop type classification
title_fullStr Galileo pretrained remote sensing model – Rwanda crop type classification
title_full_unstemmed Galileo pretrained remote sensing model – Rwanda crop type classification
title_short Galileo pretrained remote sensing model – Rwanda crop type classification
title_sort galileo pretrained remote sensing model rwanda crop type classification
topic artificial intelligence
cartography-mapping
url https://hdl.handle.net/10568/178176
work_keys_str_mv AT kenduiywobensonkipkemboi galileopretrainedremotesensingmodelrwandacroptypeclassification
AT chemuttjosephkipkoech galileopretrainedremotesensingmodelrwandacroptypeclassification