Publikation
Multi-Modal Fusion Methods with Local Neighborhood Information For Crop Yield Prediction at Field and Subfield Levels
Miro Miranda Lorenz; Deepak Kumar Pathak; Marlon Nuske; Andreas Dengel (Hrsg.)
IEEE International Geoscience and Remote Sensing Symposium (IGARSS-2024), July 7-12, Athens, Greece, IEEE, 2024.
Zusammenfassung
Yield prediction at both field and subfield level poses a significant challenge, yet it holds paramount importance for decision-making and food security within the agricultural sector. Recent efforts, focused on integrating remote sensing data coupled with machine learning models, thereby creating globally scalable models for various crop types. This study underscores the effectiveness of Sentinel-2 and complementary data sources such as weather, soil, and terrain in enhancing yield prediction. We address limitations of previous works and introduce a framework that incorporates local neighborhood information using a convolutional neural network approach. Additionally, we address the complexity of sensor fusion, showcasing both early fusion and late fusion frameworks. Notably. This study reports an R2 of 0.83 for soybean in Argentina. The results are demonstrated on a large yield dataset for Soybean, Wheat, and Rapeseed distributed across multiple countries, including Argentina, Uruguay, and Germany