Creating Open Agricultural Maps and Groundtruth Data to Better Deliver Farm Extension Services
This project aims to help improve agricultural extension services for African smallholder farmers while providing a scalable, sustainable model for generating open agricultural maps and ground-truth data. This goal will be achieved by combining: 1) an agricultural extension team equipped with an award-winning farmer engagement platform with 2) an advanced cropland mapping system. Cropland maps will help extension agents to efficiently find new customers, from whom they will collect crop type observations that will be used to develop crop type maps. The process of generating these essential agricultural data will further improve the quality and reach of extension services.
Alloysius Attah, Farmerline
Lyndon Estes, Clark University
Final Report – Phase 2
The second phase of the project addressed the challenge of collecting ground truth data while simultaneously developing relevant and timely agricultural maps. This final report provides an overview of the methods for collecting data, integrating machine learning models, and satellite images to create field boundary and crop type maps.
Final Report – Phase 1
The project was developed to address the challenges of collecting ground truth data on crop types in a sustained manner and to create reliable maps of crop types that cover large areas and multiple seasons. Read the final report to learn about the cropland maps and value-added products created for the Ejura-Tain region in Ghana.
Crop Type Maps – Lessons Learned
This project collects crop-type data in order to test its sustainably over large areas. Read the report to learn about initial lessons learned from the gathering crop type maps for Maize and Rice.
Scaling-Up Data Collection Plan
Read the Service Scale-Up plan that describes improvements the team will implement as they extend their project scope to cover additional regions in Ghana.