Early Yield Prediction for Sugar Beet Fields using Satellite Data -- Learnings from Specialized Vision Transformers
2026-07-20 • Computer Vision and Pattern Recognition
Computer Vision and Pattern RecognitionMachine Learning
AI summaryⓘ
The authors studied how to predict sugar beet harvest yields early using satellite images from the Sentinel-2 system. They combined farming knowledge with machine learning and found that using very small image pieces and all color bands helped improve their prediction model. They also created a way to identify fields likely to produce less sugar beet early in the growing season. This approach shows practical value for early detection of low-yield crops.
Remote sensingSatellite imagerySentinel-2Machine learningVision transformerSpectral bandsCrop yield forecastingAgricultural monitoringRanking-based detection
Authors
Philipp Vaeth, Bhumika Laxman Sadbhave, Denise Dejon, Gunther Schorcht, Magda Gregorova
Abstract
Remote sensing has become an increasingly valuable tool for agricultural monitoring, particularly through the use of publicly available satellite imagery. However, effectively integrating domain knowledge into machine learning methods remains challenging. This study presents a real-world example of early sugar beet harvest yield forecasting from purely optical Sentinel-2 imagery, demonstrating how a tight integration of domain knowledge and machine learning can lead to synergistic gains. We empirically find that using very small vision transformer patch sizes and all available Sentinel-2 spectral bands improves our model despite being uncommon design choices in the domain. As a practical contribution, we were able to identify a large fraction of low-yield fields in a different year early on in the growth cycle through a modified training setup and a ranking-based detection of underperforming fields.