Transforming agriculture with edge AI – enabling the Smart Farming

DOI

https://www.doi.org/10.59256/ijire.20260702030

Published In

Volume 7, Issue 2 (2026)

Published Online

March-April 2026

Pages

244-253

Article Type

Original Article

Check for updates

This project focuses on building an intelligent plant disease classification system using the Plant Village dataset from Kaggle, specifically targeting three tomato leaf categories: healthy, late blight, and bacterial spot. The dataset was cleaned, resized, and augmented, then split into an 80:20 ratio for training and validation. Model development and training were carried out in Amazon SageMaker Studio, leveraging its scalable compute environment and integrated experiment tracking. After achieving satisfactory accuracy, the trained TensorFlow model was exported and deployed directly within SageMaker using a managed inference endpoint. For accessibility, a lightweight Gradio-based user interface is being built to allow users to upload leaf images and receive instant predictions through the deployed model. The final solution demonstrates a complete machine-learning workflow—from dataset preparation to cloud deployment and user interaction—providing a practical tool for early crop disease detection and supporting precision agriculture.