Title: Lightweight CNN and blockchain integrated secured model for crop disease information system

Authors: Neenu Johnson; M.B. Santosh Kumar; T. Dhannia

Addresses: Division of Information Technology, School of Engineering, Cochin University of Science and Technology (CUSAT), Kochi, Kerala, India ' Division of Information Technology, School of Engineering, Cochin University of Science and Technology (CUSAT), Kochi, Kerala, India ' Division of Applied Sciences and Humanities, School of Engineering, Cochin University of Science and Technology (CUSAT), Kochi, Kerala, India

Abstract: Crop diseases are a major threat to the farming community that reduce yield and affect the income of farmers. Higher profitability, attaining sustainability, minimising workload, and economic development of the nation are the main driving forces for the adoption of smart farming technologies. Deep learning has emerged as an accurate tool for prediction and decision-making in smart farming. The integration of blockchain with deep learning is efficient in developing a secured data sharing framework. A crop disease data management framework that leverages the benefits of deep learning, blockchain, and interplanetary file system is proposed to assist farmers in crop disease detection and secure sharing of crop disease data. In the framework, a lightweight convolutional neural network architecture-based banana crop disease detection module is included that achieves an accuracy of 99.91%. A system architecture based on blockchain for the crop disease communication module is included to ensure secured crop data sharing.

Keywords: deep learning; blockchain; lightweight convolutional neural networks; lightweight-CNNs; agriculture; crop disease detection; hyperledger fabric; interplanetary file system; IPFS; smart contract; TensorFlow lite model; Raspberry Pi 4.

DOI: 10.1504/IJSAMI.2024.137707

International Journal of Sustainable Agricultural Management and Informatics, 2024 Vol.10 No.2, pp.162 - 184

Received: 19 Mar 2023
Accepted: 12 Jul 2023
Published online: 02 Apr 2024 *

Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article