CROP-YIELD-FORECASTING-USING-SATELLITE-IMAGERY-AND-ML-TECHNIQUES
Keywords:
Crop Yield Prediction, Machine Learning, Random Forest, LSTM, NDVI, Precision Agriculture, Time Series Forecasting.Abstract
Accurate crop yield prediction plays a critical role in agricultural planning, food security, and economic stability. This paper presents a machine learning and deep learning-based crop yield prediction system that integrates satellite vegetation indices and meteorological parameters. The proposed system utilizes Random Forest (RF) and Long Short-Term Memory (LSTM) models to analyze normalized difference vegetation index (NDVI), enhanced vegetation index (EVI), rainfall, temperature, and humidity data. The dataset consists of historical agricultural yield records combined with weather and satellite-derived features. Experimental results demonstrate that ensemble learning and time-series modeling significantly improve prediction accuracy. The LSTM model effectively captures temporal dependencies, while Random Forest provides strong baseline regression performance. The proposed approach can support precision agriculture and data-driven decision-making.
