From space to soil: Advancing crop mapping and ecosystem insights for smallholder agriculture

This project centers on in-season crop type mapping in Nandi County, Kenya, utilizing time-series Sentinel-2 imagery and supervised machine learning techniques. The objective is to produce accurate crop-type maps to support agricultural management activities such as yield estimation, acreage statist...

Descripción completa

Detalles Bibliográficos
Autor principal: Guo, Zhe
Formato: Brief
Lenguaje:Inglés
Publicado: CGIAR System Organization 2024
Materias:
Acceso en línea:https://hdl.handle.net/10568/168470
Descripción
Sumario:This project centers on in-season crop type mapping in Nandi County, Kenya, utilizing time-series Sentinel-2 imagery and supervised machine learning techniques. The objective is to produce accurate crop-type maps to support agricultural management activities such as yield estimation, acreage statistics, disaster damage assessment, and ecosystem evaluation. The approach leverages cloud-based computing, offering a customized and flexible solution that requires no prior knowledge of cloud infrastructure.