Tytuł pozycji:
Binary Classification of Agricultural Crops Using Sentinel Satellite Data and Machine Learning Techniques
The advent of high-resolution satellite imagery, such as Sentinel 1 and Sentinel 2, has provided valuable data for various applications, including crop classification. This paper presents a study on the classification of agricultural fields using indices derived from Sentinel satellite imagery. Specifically, we focus on creating binary classifiers capable of distinguishing between different crops, namely Tomatoes, Soy, Sugar Beets, Rice, and Wheat. The paper investigates various preprocessing techniques to create a dataset suitable for machine learning methods, such as Random Forests, which require a fixed number of features. Additionally, we demonstrate that linear interpolation and out-of-scale values have equivalent performance in terms of classification accuracy. Furthermore, we address the issue of imbalanced datasets commonly encountered in agricultural field classification. We explore different balancing techniques that can significantly improve the performance of machine learning methodologies. The motivation for this work stems from the growing interest in Agriculture 4.0, and it serves as a valuable tool to verify farmers' claims, especially in relation to state subsidies for specific crops of interest. Understanding the crop type present in the field represents highly valuable information that can serve as a foundation for subsequent analyses or as input for calibrating models, such as Decision Support Systems. Overall, this study contributes to the field of agricultural research and provides insights into the application of Machine Learning techniques for crop classification using satellite imagery. The findings offer practical implications for monitoring and optimizing agricultural practices in the context of precision farming and sustainable agriculture.
1. This work has been partially supported by the INdAM research group GNCS. The publication was created with the co-financing of the European Union-FSE-REACT-EU, PON Research and Innovation 2014-2020 DM1062/2021. This work has been partially supported by the industrial collaboration between ABACO Group SpA and UNIMORE.
2. Thematic Tracks Short Papers
3. Opracowanie rekordu ze środków MEiN, umowa nr SONP/SP/546092/2022 w ramach programu "Społeczna odpowiedzialność nauki" - moduł: Popularyzacja nauki i promocja sportu (2024).
