Iot-based AI prediction system for early and late blight disease in tomato crops

Overview

This solution addresses a critical challenge in tomato farming: the unpredictable onset of early and late blight disease, which can cause significant crop losses. The system integrates an IoT-based sensor module with artificial intelligence classifiers (Artificial Neural Networks and Support Vector Machines) to predict blight occurrence at each individual growth stage of the tomato crop.

By deploying sensors directly at the site of investigation, the system captures accurate, real-time weather and soil parameters that feed into predictive models. Results are delivered through a mobile application that provides growers with actionable blight probability information specific to the current growth stage of their crop. The modular design allows the same sensor and prediction framework to be customized for other crops and agricultural research needs.

Technical specifications

• IoT sensor module for continuous monitoring of weather and soil parameters at the farm site, enabling more accurate data collection than remote or secondary sources
• Stage-specific predictive models built using ANN and SVM classifiers, trained to assess blight risk for each individual growth stage of the tomato crop
• Real-time data acquisition and storage for ongoing environmental monitoring and historical analysis
• Mobile application interface that presents blight onset probability predictions based on current weather and soil conditions
• Customizable sensor configuration allowing the module to be adapted to different sites and crop research requirements
• Scalable architecture suitable for deployment across multiple farmlands for broader agricultural monitoring

Technology readiness level

The proposed solution is at an early-to-mid stage of development. Current blight prediction research relies on secondary data sources; this project aims to advance the field by introducing on-site IoT measurement for improved accuracy. Future validation will involve deploying the IoT device on farmland to continuously capture environmental data and verify the accuracy of blight predictions. The expected outcomes include a robust sensor module, a validated AI prediction tool delivered via app, and a customizable platform extendable to additional agricultural research applications.