Enhancing Food Security With High-Quality Land-Use and Land-Cover Maps: A Local Model Approach
In 2023, 58% of the African population experienced moderate to severe food insecurity, with 21.6% facing severe food insecurity. Land-use and land-cover maps enable informed resource management, urban planning, and environment monitoring to enhance food security. The development of global land-cover...
Saved in:
| Main Authors: | , , , , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
IEEE
2025-01-01
|
| Series: | IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing |
| Subjects: | |
| Online Access: | https://ieeexplore.ieee.org/document/11008699/ |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1849430216543305728 |
|---|---|
| author | Girmaw Abebe Tadesse Caleb Robinson Charles Mwangi Esther Maina Joshua Nyakundi Luana Marotti Gilles Quentin Hacheme Hamed Alemohammad Rahul Dodhia Juan M. Lavista Ferres |
| author_facet | Girmaw Abebe Tadesse Caleb Robinson Charles Mwangi Esther Maina Joshua Nyakundi Luana Marotti Gilles Quentin Hacheme Hamed Alemohammad Rahul Dodhia Juan M. Lavista Ferres |
| author_sort | Girmaw Abebe Tadesse |
| collection | DOAJ |
| description | In 2023, 58% of the African population experienced moderate to severe food insecurity, with 21.6% facing severe food insecurity. Land-use and land-cover maps enable informed resource management, urban planning, and environment monitoring to enhance food security. The development of global land-cover maps has been facilitated by the increasing availability of earth observation data and advancements in geospatial machine learning. However, these global maps exhibit lower accuracy and inconsistencies in Africa, partly due to the lack of representative training data. To address this issue, we propose a data-centric framework with a teacher–student model setup, which uses diverse data sources of satellite images and label examples to produce local land-cover maps. Our method trains a high-resolution teacher model on images with a resolution of 0.331 <inline-formula><tex-math notation="LaTeX">$\mathtt {m/pixel}$</tex-math></inline-formula> and a low-resolution student model on publicly available images with a resolution of 10 <inline-formula><tex-math notation="LaTeX">$\mathtt {m/pixel}$</tex-math></inline-formula>. The student model also utilizes the teacher model’s output as its weak label examples as a form of outcome-based knowledge distillation. We evaluated our framework using Murang’a county in Kenya, renowned for its agricultural productivity, as a use case. Our local models achieved higher quality maps, with improvements of 0.14 in the <inline-formula><tex-math notation="LaTeX">$F_{1}$</tex-math></inline-formula> score and 0.21 in Intersection-over-Union, compared to the best global model. Our evaluation also revealed inconsistencies in existing global maps, with a maximum agreement rate of 0.30 among themselves. Our work provides valuable guidance to decision-makers for driving informed decisions to enhance food security. |
| format | Article |
| id | doaj-art-9feb8d4ee3c242ee810e96919e71d01e |
| institution | Kabale University |
| issn | 1939-1404 2151-1535 |
| language | English |
| publishDate | 2025-01-01 |
| publisher | IEEE |
| record_format | Article |
| series | IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing |
| spelling | doaj-art-9feb8d4ee3c242ee810e96919e71d01e2025-08-20T03:28:05ZengIEEEIEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing1939-14042151-15352025-01-0118152651527710.1109/JSTARS.2025.357224711008699Enhancing Food Security With High-Quality Land-Use and Land-Cover Maps: A Local Model ApproachGirmaw Abebe Tadesse0https://orcid.org/0000-0002-2648-9102Caleb Robinson1https://orcid.org/0000-0003-1975-4454Charles Mwangi2Esther Maina3Joshua Nyakundi4Luana Marotti5Gilles Quentin Hacheme6https://orcid.org/0000-0002-9465-6558Hamed Alemohammad7https://orcid.org/0000-0001-5662-3643Rahul Dodhia8Juan M. Lavista Ferres9https://orcid.org/0000-0002-9654-3178Microsoft AI for Good Research Lab, Redmond, WA, USAMicrosoft AI for Good Research Lab, Redmond, WA, USAKenya Space Agency, Nairobi, KenyaKenya Space Agency, Nairobi, KenyaKenya Space Agency, Nairobi, KenyaMicrosoft AI for Good Research Lab, Redmond, WA, USAMicrosoft AI for Good Research Lab, Redmond, WA, USAClark University, Worcester, MA, USAMicrosoft AI for Good Research Lab, Redmond, WA, USAMicrosoft AI for Good Research Lab, Redmond, WA, USAIn 2023, 58% of the African population experienced moderate to severe food insecurity, with 21.6% facing severe food insecurity. Land-use and land-cover maps enable informed resource management, urban planning, and environment monitoring to enhance food security. The development of global land-cover maps has been facilitated by the increasing availability of earth observation data and advancements in geospatial machine learning. However, these global maps exhibit lower accuracy and inconsistencies in Africa, partly due to the lack of representative training data. To address this issue, we propose a data-centric framework with a teacher–student model setup, which uses diverse data sources of satellite images and label examples to produce local land-cover maps. Our method trains a high-resolution teacher model on images with a resolution of 0.331 <inline-formula><tex-math notation="LaTeX">$\mathtt {m/pixel}$</tex-math></inline-formula> and a low-resolution student model on publicly available images with a resolution of 10 <inline-formula><tex-math notation="LaTeX">$\mathtt {m/pixel}$</tex-math></inline-formula>. The student model also utilizes the teacher model’s output as its weak label examples as a form of outcome-based knowledge distillation. We evaluated our framework using Murang’a county in Kenya, renowned for its agricultural productivity, as a use case. Our local models achieved higher quality maps, with improvements of 0.14 in the <inline-formula><tex-math notation="LaTeX">$F_{1}$</tex-math></inline-formula> score and 0.21 in Intersection-over-Union, compared to the best global model. Our evaluation also revealed inconsistencies in existing global maps, with a maximum agreement rate of 0.30 among themselves. Our work provides valuable guidance to decision-makers for driving informed decisions to enhance food security.https://ieeexplore.ieee.org/document/11008699/AIafricaagriculturefood securityland cover |
| spellingShingle | Girmaw Abebe Tadesse Caleb Robinson Charles Mwangi Esther Maina Joshua Nyakundi Luana Marotti Gilles Quentin Hacheme Hamed Alemohammad Rahul Dodhia Juan M. Lavista Ferres Enhancing Food Security With High-Quality Land-Use and Land-Cover Maps: A Local Model Approach IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing AI africa agriculture food security land cover |
| title | Enhancing Food Security With High-Quality Land-Use and Land-Cover Maps: A Local Model Approach |
| title_full | Enhancing Food Security With High-Quality Land-Use and Land-Cover Maps: A Local Model Approach |
| title_fullStr | Enhancing Food Security With High-Quality Land-Use and Land-Cover Maps: A Local Model Approach |
| title_full_unstemmed | Enhancing Food Security With High-Quality Land-Use and Land-Cover Maps: A Local Model Approach |
| title_short | Enhancing Food Security With High-Quality Land-Use and Land-Cover Maps: A Local Model Approach |
| title_sort | enhancing food security with high quality land use and land cover maps a local model approach |
| topic | AI africa agriculture food security land cover |
| url | https://ieeexplore.ieee.org/document/11008699/ |
| work_keys_str_mv | AT girmawabebetadesse enhancingfoodsecuritywithhighqualitylanduseandlandcovermapsalocalmodelapproach AT calebrobinson enhancingfoodsecuritywithhighqualitylanduseandlandcovermapsalocalmodelapproach AT charlesmwangi enhancingfoodsecuritywithhighqualitylanduseandlandcovermapsalocalmodelapproach AT esthermaina enhancingfoodsecuritywithhighqualitylanduseandlandcovermapsalocalmodelapproach AT joshuanyakundi enhancingfoodsecuritywithhighqualitylanduseandlandcovermapsalocalmodelapproach AT luanamarotti enhancingfoodsecuritywithhighqualitylanduseandlandcovermapsalocalmodelapproach AT gillesquentinhacheme enhancingfoodsecuritywithhighqualitylanduseandlandcovermapsalocalmodelapproach AT hamedalemohammad enhancingfoodsecuritywithhighqualitylanduseandlandcovermapsalocalmodelapproach AT rahuldodhia enhancingfoodsecuritywithhighqualitylanduseandlandcovermapsalocalmodelapproach AT juanmlavistaferres enhancingfoodsecuritywithhighqualitylanduseandlandcovermapsalocalmodelapproach |