Estimating Crop Yield Utilizing Machine Learning And Sentinel Photography

View at publisher Summarise with VIRAT
Authors: Gagan Jain, Manan Gohil, Mujtaba Mohamed, Nasim Banu Shah
Cited by: 0

Summary

No summary yet. VIRAT can read this paper and write a plain-language summary of its aim, methods, key findings and limitations.

Summarise with VIRAT Free to try with your Science Catalogs account.

Abstract

Crop yield prediction is extremely pivotal in addressing global food security concerns and effective management of farm resources. The current study analyzes the complementarity of Sentinel-2 multispectral Machine learning models and high-resolution data in crop yield estimation for Karnataka, India, during 2016–2024. Use of vegetation indices such as Normalized Difference Vegetation Index (NDVI) and Normalized Difference Water Index (NDWI) and meteorological variables such as precipitation, temperature, humidity, and cloud cover led to the construction of a Random Forest model to evaluate the complex interactions of environmental variables with farm productivity. The model performed exceptionally well with prediction R² = 0.78 – 0.85, Mean Absolute Error (MAE) = 0.25 – 0.35 t/ha, and Root Mean Squared Error (RMSE) = 0.32 – 0.45 t/ha. The observations reflect the efficacy of integrating remote sensing data with meteorological variables for further improved predictions. The findings of this research provide useful information for precision agriculture, enabling policymakers and farmers to make efficient decisions toward improving farm productivity and sustainable agriculture. Future work can further improve these predictions using additional satellite datasets and advanced deep learning tools.

Fields of Study & Subjects

Remote Sensing in Agriculture Smart Agriculture and AI Climate change impacts on agriculture

Publication details

Journal / source
CRC Press eBooks
Pages
246–252
Publication date
25 March 2026
Language
EN
Document type
Book chapter