SOIL YIELD FORECASTING

This research project serves as a comprehensive meta-analysis in the field of agricultural science, specifically focusing on the prediction of crop yields. This endeavor involves collating and synthesizing findings from a variety of studies and articles that have explored different methodologies and models for forecasting agricultural outputs. The objective of this comprehensive review is to identify trends, methodologies, and key factors that consistently influence crop yield predictions across different studies.
It synthesizes methodologies from various studies, emphasizing machine learning (ML) techniques like Support Vector Machines (SVM), Random Forest (RF), and Convolutional Neural Networks (CNN). These studies integrate high-resolution satellite imagery with environmental indices such as NDVI, EVI, and LAI. Soil chemical properties (pH, nutrients) and satellite-derived data were used to enhance the prediction of crop yields for diverse crops. The findings highlight the comparative effectiveness of different models in handling the spatial and temporal variability of both above-ground and below-ground data, improving prediction accuracy under varying environmental and soil conditions.
Through this theoretical analysis, the research underscores the potential of advanced analytical models to transform agricultural monitoring and prediction, providing critical insights that can aid in the optimization of agricultural policies and resource management.

1. Statistical Estimation of Crop Yields for the Midwestern United States Using Satellite Images, Climate Datasets, and Soil Property Maps / N. Kim et al // Korean Journal of Remote Sensing. – 2016. – № 32(4). – Р. 383-401.

2. Kim N. Machine Learning Approaches to Corn Yield Estimation Using Satellite Images and Climate Data: A Case of Iowa State / N. Kim, Y.-W. Lee // Journal of the Korean Society of Surveying, Geodesy, Photogrammetry and Cartography. – 2016. – № 34(4). – Р. 383-390.

3. Chang J. Identifying Factors for Corn Yield Prediction Models and Evaluating Model Selection Methods / J. Chang, D.E. Clay // Korean Journal of Crop Science. – 2005. – № 50(4). – Р. 268-275.

4. Boosted random forest / Y. Mishina et al // IEICE Transactions on Information and Systems. – 2005. – № E98.D(9). – Р. 1630-1636. https://doi.org/10.1587/transinf.2014OPP0004.

5. Li.Y. Regional segmentation of field images based on convolutional neural network for rice combine harvester. / Y. Li., M. Iida, M. Suguri, R. Masuda // Journal of the Japanese Society of Agricultural Machinery and Food Engineers. – 2020. – № 82(1). – Р. 47-56. https://doi.org/10.11357/jsamfe.82.1_47.

6. Park H. Satellite-based cabbage and radish yield prediction using deep learning in Kangwon-do / H. Park, Y. Lee, S. Park // Korean Journal of Remote Sensing. – 2023. – Vol. 39. – Р. 1031-1042. https://koreascience.kr/article/JAKO202331857673593.page.

7. Sung J.H. Rice yield prediction based on the soil chemical properties using neural network model / J.H. Sung, D.H. Lee // Journal of Biosystems Engineering. – 2004. – № 29(2). – Р. 123-135.

8. Relationships between global climate indices and rain-fed crop yields in highland of SouthCentral Java, Indonesia / B.D.A. Nugroho et al // Journal of Geography (Chigaku Zasshi). – 2013. – № 122(3). – Р. 438-447. https://doi.org/10.5026/jgeography.122.438.

9. Purwanto M.Y.J. Crop yield prediction by stress day indices under both excessive and deficient soil water conditions / M.Y.J. Purwanto, S. Hardjoamidjojo, R. Nakamura, N. Kubo // Journal of Irrigation Engineering and Rural Planning. – 1993. – № 25. – Р. 31-41. https://doi.org/10.11408/jierp1982.1993.25_31.

10. Jun S.H. A differential evolution based support vector clustering / S.H. Jun // Journal of the Korean Institute of Intelligent Systems. – 2007. – № 17(5). – Р. 679-683. https://doi.org/10.5391/JKIIS.2007.17.5.679.