IoT-Driven Smart Crop Health Monitoring and Advisory System

Abstract

Even in today’s digital age, Indian agriculture still largely relies on traditional or outdated farming practices that have been passed down through generations, with limited integration of data-driven and intelligent solutions for efficient and sustainable farming. To overcome this limitation, we propose a Smart Agriculture System that integrates agriculture, IoT technologies, and Machine Learning. The smart system allows users to observe soil nutrients (NPK), temperature, humidity, and moisture levels using the IoT sensors in real-time and to capture and upload leaf images via the mobile application, which are then analysed to detect and diagnose leaf diseases with a confidence score. By combining these inputs, the system generates results such as overall soil condition status and fertilizer recommendations. These results are then showed in the mobile application, helping farmers make well-versed decisions, improve their resource usage, rise crop yields, and promote efficient, sustainable, and environment friendly farming practices.

Introduction

The proposed IoT-Driven Crop Health Monitoring and Advisory System is a smart agriculture platform that combines Internet of Things (IoT), Artificial Intelligence (AI), Machine Learning (ML), mobile technology, and cloud computing to improve farming efficiency, productivity, and sustainability. The system enables farmers to monitor soil conditions, detect plant diseases, and receive real-time recommendations for fertilizer and irrigation management through a mobile application.

The system integrates multiple components, including IoT-based soil monitoring, image-based disease detection, and an intelligent advisory system. Soil parameters such as moisture, temperature, humidity, and NPK levels are collected using an ESP32-based sensor module, while leaf images are analyzed using a CNN-based machine learning model developed with TensorFlow/Keras and OpenCV. The collected data is stored and synchronized using Firebase Realtime Database, while a Flutter-based mobile application provides users with an easy-to-use interface for monitoring, diagnosis, and recommendations.

The main objectives of the project are to develop an integrated smart farming solution that can automatically detect crop diseases, monitor environmental conditions, analyze plant health, generate recommendations, reduce resource wastage, and support sustainable farming practices. The system is designed to be affordable and accessible, especially for small and medium-scale farmers.

The methodology follows a multi-layer architecture consisting of a user interface layer, application processing layer, database layer, machine learning layer, and IoT layer. The development process includes image acquisition, sensor data collection, ML model training, data preprocessing, backend processing, mobile application development, cloud integration, testing, and validation. The CNN model is trained using healthy and diseased leaf image datasets with augmentation techniques such as rotation, flipping, and scaling to improve accuracy and reliability.

The developed system successfully integrates mobile application features, IoT sensing, machine learning analysis, and cloud services into a single platform. The diagnosis module identifies plant diseases from leaf images and provides confidence scores along with crop health trends. The advisory module analyzes disease results, soil nutrients, temperature, and humidity data to provide recommendations for fertilizer usage, irrigation, and preventive measures. The assistant module offers user-friendly explanations and guidance to help farmers understand system outputs and make better decisions.

Conclusion

The IoT-Driven Smart Crop Health Monitoring and Advisory System combines machine learning and IoT to improve modern farming. It integrates image-based disease detection with real-time monitoring of temperature, humidity, and NPK levels to provide accurate suggestions. A CNN model identifies leaf diseases, while sensor data is processed to generate fertilizer and irrigation recommendations. With Firebase enabling real-time communication, the system enhances decision-making, reduces wastage of resources, and promotes efficient and sustainable agricultural practices.

References

[1] IoT-Based Systems for Soil Nutrients Assessment in Horticulture. Available at: https://www.mdpi.com/1424-8220/23/1/403 [2] Machine Learning based soil moisture prediction for Internet of Things based Smart Irrigation System. Available at: https://ieeexplore.ieee.org/abstract/document/8988313 [3] IoT-Enabled Soil Nutrient Analysis and Crop Recommendation Model for Precision Agriculture. Available at: https://www.mdpi.com/2073-431X/12/3/61 [4] Real-Time Plant Disease Dataset Development and Detection of Plant Disease Using Deep Learning. Available at: https://www.researchgate.net/publication/377707771_Real-time_Plant_Disease_Dataset_Development_and_Detection_of_Plant_Disease_Using_Deep_Learning [5] Improving Plant Disease Classification With Deep-Learning-Based Prediction Model Using Explainable Artificial Intelligence j-Hassan, M., Chaker, R., and Cederqvist, A.-M. (2024), Environmental Education: A Systematic Review on the Use of Digital Tools for Fostering Sustainability Awareness, Sustainability, 16(9), 3733. Available at: https://www.researchgate.net/publication/382282535_Improving_Plant_Disease_Classification_with_Deep_Learning_based_Prediction_Model_using_Explainable_Artificial_Intelligence [6] Soil Nutrient Prediction and Crop Recommendation System: Conventional to Modern Methods of Soil NPK Sensing. Available at: https://ieeexplore.ieee.org/document/10915247 [7] Enhancing Crop Yield and Growth Prediction Using IoT-Based Sma Irrigation Systems and Machine Learning Algorithms. Available at: https://ieeexplore.ieee.org/document/10698825 [8] Enhancing Personalized Agriculture with Cutting-Edge Hybrid Machine Learning Models and Intelligent Edge Computing for Sustainable Farming. Available at: https://www.researchgate.net/publication/392225412_Enhancing_Personalized_Agriculture_With_Cutting-Edge_Hybrid_Machine_Learning_Models_And_Intelligent_Edge_Computing_For_Sustainable_Farming [9] Plant Disease Detection in Precision Agriculture: Deep Learning Approaches. Available at: https://ieeexplore.ieee.org/document/10467525 [10] Analysis of Internet of Things based Agriculture Fertilizer Nutrient Management Soil Health Irrigation System and its Applications. Available at: https://ieeexplore.ieee.org/document/10467345

Copyright

Copyright © 2026 K S S Pranathi, V Sai Srinidhi, B Sneha, Maya B Dhone. This is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.