Blockchain based precision rice farming framework using deep learning techniques

Abstract

Food security is a critical issue, and rice is one of the world’s staple foods on which more than half of the world feeds. Its growing demand calls for sustainable agriculture methods that can meet present and future demands while tackling resource restrictions and climate change. Precision agriculture, or smart agriculture (SA), uses data analytics, sensors, drones, and machine learning (ML) algorithms to enhance agricultural techniques, minimise waste, and alleviate environmental effects. This study seeks to provide a safe framework (RiceBlock model) for precision rice farming integrating blockchain, Internet of Things (IoT), and deep learning technologies. The framework proposes an IoT-enabled sensor network for real-time agricultural data collection, a blockchain for securing the collected data, and a deep learning-driven analytics model for rice yield prediction and automated decision-making. The proposed model demonstrated superior rice yield prediction accuracy against the state-of-the-art models in terms of R2 (0.97), RMSE (1.38), MAE (1.16), and NRMSE (0.04). It ensures data security, integrity, and resilience against known attacks and data manipulation.