Winter wheat yield prediction using UAV imagery and machine learning: case studies in Sweden and Morocco
(2025) In Student thesis series INES NGEM01 20251Dept of Physical Geography and Ecosystem Science
- Abstract
- To address global food security challenges caused by population growth and climate change, accurate prediction of crop yield is essential for sustainable agriculture. Winter wheat, a key global crop, faces production instability in regions like Sweden and Morocco due to extreme weather conditions. This study employed unmanned aerial vehicle (UAV)-based multispectral imagery and digital surface models (DSMs) data derived from UAV data, integrated with three machine learning models, Random Forest (RF), Support Vector Machine Regression (SVR), and Extreme Gradient Boosting (XGBoost), to forecast winter wheat yield at high spatial resolution in experimental fields in Sweden (humid climate) and Morocco (arid to semi-arid climate). Prediction... (More)
- To address global food security challenges caused by population growth and climate change, accurate prediction of crop yield is essential for sustainable agriculture. Winter wheat, a key global crop, faces production instability in regions like Sweden and Morocco due to extreme weather conditions. This study employed unmanned aerial vehicle (UAV)-based multispectral imagery and digital surface models (DSMs) data derived from UAV data, integrated with three machine learning models, Random Forest (RF), Support Vector Machine Regression (SVR), and Extreme Gradient Boosting (XGBoost), to forecast winter wheat yield at high spatial resolution in experimental fields in Sweden (humid climate) and Morocco (arid to semi-arid climate). Prediction results revealed that the grain-filling growth stage was the best stage for wheat yield prediction in both sites, with the SVR model demonstrating best performance (Sweden: R²=0.88, MAE=0.71 t/ha, RMSE=0.96 t/ha; Morocco: R²=0.83, MAE=0.42 t/ha, RMSE=0.56 t/ha). Four spectral band reflectance and ten vegetation indices (VIs) showed strong correlations with yield during the heading/flowering and grain-filling stages in both regions. Combing relative wheat height information enhanced prediction accuracy during early growth stages (jointing /booting stage and heading/flowering stage) but introduced uncertainty in the grain-filling stage. Cross-regional model transferability was limited, with better results when using the larger Swedish dataset to predict yields in Morocco than vice versa, underscoring the role of dataset size and yield variability. These results confirm the potential of UAV-based remote sensing combined with machine learning for precise, within-field winter wheat yield predictions, providing practical insights for improving agricultural strategies. The study emphasizes the importance of phenological timing, spectral features, and sufficient data volume, laying a foundation for future advancing cross-regional yield prediction methodologies. (Less)
- Popular Abstract
- Under the context of a global growing population and climate change impact, accurately predict crop yields is more and more crucial than ever. This study focuses on winter wheat, one of the key global crop, and explores how drone images and artificial intelligence(AI) methods can improve yield prediction in two different environments: Sweden(humid) and Morocco(semi-arid).
High-resolution images were collected using drones during different wheat growth stages. Three machine learning models—Random Forest, Support Vector Machine Regression (SVR), and Extreme Gradient Boosting—were used to analyze the data and predict wheat yields. Results showed that the grain-filling stage was the most effective time for prediction. Among all models, SVR... (More) - Under the context of a global growing population and climate change impact, accurately predict crop yields is more and more crucial than ever. This study focuses on winter wheat, one of the key global crop, and explores how drone images and artificial intelligence(AI) methods can improve yield prediction in two different environments: Sweden(humid) and Morocco(semi-arid).
High-resolution images were collected using drones during different wheat growth stages. Three machine learning models—Random Forest, Support Vector Machine Regression (SVR), and Extreme Gradient Boosting—were used to analyze the data and predict wheat yields. Results showed that the grain-filling stage was the most effective time for prediction. Among all models, SVR performed best, achieving 88% accuracy in Sweden and 83% in Morocco. Key indicators like vegetation indices and spectral reflectance were strongly linked to yield, especially during flowering and grain-filling stages. Including wheat height improved early-stage predictions but was less reliable later. Cross-regional predictions were more successful when using the larger Swedish dataset to predict yields in Morocco, highlighting the importance of data volume and variability.
This study shows the strong potential of combining drone technology with AI to make precise, field-level yield forecasts, helping farmers make better decisions and contributing to more sustainable agriculture. (Less)
Please use this url to cite or link to this publication:
http://lup.lub.lu.se/student-papers/record/9206553
- author
- Yu, Mengjie LU
- supervisor
- organization
- course
- NGEM01 20251
- year
- 2025
- type
- H2 - Master's Degree (Two Years)
- subject
- keywords
- winter wheat, yield predictions, machine learning, UAV multispectral imagery, field scale, Sweden, Morocco
- publication/series
- Student thesis series INES
- report number
- 745
- funder
- Crafoord Foundation
- language
- English
- id
- 9206553
- date added to LUP
- 2025-06-27 14:44:43
- date last changed
- 2025-06-27 14:44:43
@misc{9206553,
abstract = {{To address global food security challenges caused by population growth and climate change, accurate prediction of crop yield is essential for sustainable agriculture. Winter wheat, a key global crop, faces production instability in regions like Sweden and Morocco due to extreme weather conditions. This study employed unmanned aerial vehicle (UAV)-based multispectral imagery and digital surface models (DSMs) data derived from UAV data, integrated with three machine learning models, Random Forest (RF), Support Vector Machine Regression (SVR), and Extreme Gradient Boosting (XGBoost), to forecast winter wheat yield at high spatial resolution in experimental fields in Sweden (humid climate) and Morocco (arid to semi-arid climate). Prediction results revealed that the grain-filling growth stage was the best stage for wheat yield prediction in both sites, with the SVR model demonstrating best performance (Sweden: R²=0.88, MAE=0.71 t/ha, RMSE=0.96 t/ha; Morocco: R²=0.83, MAE=0.42 t/ha, RMSE=0.56 t/ha). Four spectral band reflectance and ten vegetation indices (VIs) showed strong correlations with yield during the heading/flowering and grain-filling stages in both regions. Combing relative wheat height information enhanced prediction accuracy during early growth stages (jointing /booting stage and heading/flowering stage) but introduced uncertainty in the grain-filling stage. Cross-regional model transferability was limited, with better results when using the larger Swedish dataset to predict yields in Morocco than vice versa, underscoring the role of dataset size and yield variability. These results confirm the potential of UAV-based remote sensing combined with machine learning for precise, within-field winter wheat yield predictions, providing practical insights for improving agricultural strategies. The study emphasizes the importance of phenological timing, spectral features, and sufficient data volume, laying a foundation for future advancing cross-regional yield prediction methodologies.}},
author = {{Yu, Mengjie}},
language = {{eng}},
note = {{Student Paper}},
series = {{Student thesis series INES}},
title = {{Winter wheat yield prediction using UAV imagery and machine learning: case studies in Sweden and Morocco}},
year = {{2025}},
}