Smart AI Driven Crop Disease Prediction and Management System

Abstract

Agriculture is essential in developing countries but faces challenges like crop diseases, low yields, and poor resource management. These issues often arise from limited access to timely information and expert advice. An AI-based system addresses these problems by helping farmers make informed, data-driven decisions. It detects crop diseases from leaf images using deep learning models like CNN and MobileNetV2. It also recommends suitable crops based on environmental factors such as soil nutrients, pH, temperature, and rainfall. Additionally, it provides personalized fertilizer suggestions based on soil and crop needs. Built with Python and Flask, the system uses real agricultural datasets and features a simple web interface. Docker is used for easy deployment and scalability. This integrated solution promotes sustainable farming and reduces guesswork and costs. It aims to improve productivity and support rural farmers with accessible, AI-powered tools.

Introduction

The text describes an AI-driven agricultural system designed to help farmers overcome challenges such as crop diseases, soil degradation, and unpredictable environmental conditions, which often lead to low productivity and financial instability. Traditional farming methods based on experience are no longer sufficient, making artificial intelligence (AI) and machine learning (ML) important tools for modern agriculture.

Role of AI in Agriculture

AI enables farmers to:

• Detect crop diseases early using image-based deep learning models
• Recommend suitable crops based on soil and climate conditions
• Optimize fertilizer usage for better soil health

These technologies improve decision-making, reduce manual effort, and increase farming efficiency, especially in rural areas where expert support is limited.

Project Overview

The proposed system, an AI-Driven Crop Disease Prediction and Management System (AgroAI), aims to:

• Improve agricultural productivity
• Reduce crop losses
• Support farmers with real-time, data-driven recommendations

It integrates:

• Crop disease detection using deep learning models like MobileNetV2 trained on leaf image datasets
• Crop recommendation using soil nutrients (NPK) and weather data
• Fertilizer suggestion based on soil composition and crop requirements

Literature Review Insights

Existing research shows strong progress in:

• CNN-based disease detection with high accuracy (over 98%)
• ML-based crop recommendation systems using environmental data
• Fertilizer recommendation systems combining rule-based and ML approaches

However, most systems suffer from limitations such as:

• Dependence on high-quality or real-time data
• Lack of integration between different agricultural tasks
• Limited usability for non-technical rural users
• Poor support for localized conditions

This highlights the need for a unified, user-friendly system.

Proposed Methodology

The system is designed as a modular web-based platform with three main components:

1. Crop disease detection
2. Crop recommendation
3. Fertilizer recommendation

Architecture

• Uses a client-server model
• Frontend collects user inputs (soil data, images)
• Backend processes data using AI models and returns predictions
• Built using Python, Flask, TensorFlow, and scikit-learn with a Bootstrap-based interface

Conclusion

Conclusion

The AI-Driven Crop Disease Prediction and Management System (AgroAI) was developed with the aim of solving some of the most persistent and critical challenges faced by farmers: deciding what crop to plant, how to manage fertilizers efficiently, and how to detect crop diseases early. Through the integration of machine learning and deep learning models into a simple, accessible web-based platform, this project has demonstrated that artificial intelligence can be made both practical and impactful in the context of modern agriculture. Over the course of development, the system successfully brought together three powerful modules crop recommendation, fertilizer suggestion, and image-based disease detection each powered by well-trained models using real-world datasets. The results have shown high levels of accuracy, relevance, and usability. The web interface ensures that the system remains farmer-friendly, while the backend architecture ensures flexibility and scalability for various deployment environments, from local machines to cloud- based servers via Docker. One of the major achievements of AgroAI is its modular and extensible design, which makes it easy to improve or expand individual modules without needing to rebuild the entire system. The overall solution is lightweight, cost-effective, and practical for real-world deployment, especially in resource-constrained or rural settings where such support tools are most needed. More importantly, AgroAI is a proof-of-concept for how emerging technologies can be applied to create smart farming solutions that go beyond just predictions offering guidance, reducing risk, and improving productivity in agriculture. It embodies the shift from intuition-based to data-driven farming and encourages the digital transformation of agricultural practices.

References

[1] S. Abhishek and D. K. Vishwakarma, \"A survey on plant leaf disease detection using image processing techniques,\" International Journal of Computer Applications, vol. 975, 2020. Available: https://doi.org/10.5120/ijca2020919975 [2] M. Brahimi, K. Boukhalfa, and A. Moussaoui, \"Deep learning for tomato diseases: Classification and symptoms visualization,\" Applied Artificial Intelligence, vol. 31, no. 4, pp. 299–315, 2017. Available: https://doi.org/10.1080/08839514.2017.1315516 [3] F. Chollet, \"Keras: Deep Learning Library for Theano and TensorFlow,\" GitHub Repository, 2015. Available: https://github.com/keras-team/keras [4] Docker, \"What is a Container?\" Docker Resources, [Online]. Available: https://www.docker.com/resources/what-container [5] K. P. Ferentinos, \"Deep learning models for plant disease detection and diagnosis,\" Computers and Electronics in Agriculture, vol. 145, pp. 311–318, 2018. [Online]. Available: https://doi.org/10.1016/j.compag.2018.01.009 [6] Flask Documentation, \"Flask: Web Development, One Drop at a Time,\" Flask Official Site. [Online]. Available: https://flask.palletsprojects.com/ [7] S. Ghosal et al., \"An explainable deep machine vision framework for plant stress phenotyping,\" Proceedings of the National Academy of Sciences, vol. 115, no. 18, pp. 4613–4618, 2018. [Online]. Available: https://doi.org/10.1073/pnas.1716999115 [8] A. G. Howard et al., \"MobileNets: Efficient Convolutional Neural Networks for Mobile Vision Applications,\" arXiv preprint arXiv:1704.04861, 2017. [Online]. Available: https://arxiv.org/abs/1704.04861 [9] Indian Council of Agricultural Research (ICAR), \"Fertilizer Use and Soil Health Guidelines,\" 2021. [Online]. Available: https://icar.org.in [10] S. P. Mohanty, D. P. Hughes, and M. Salathé, \"Using deep learning for image-based plant disease detection,\" Frontiers in Plant Science, vol. 7, p. 1419, 2016. Available: https://doi.org/10.3389/fpls.2016.01419 [11] F. Pedregosa et al., \"Scikit-learn: Machine learning in Python,\" Journal of Machine Learning Research, vol. 12, pp. 2825–2830, 2011. [Online]. Available: https://www.jmlr.org/papers/volume12/pedregosa11a/pedregosa11a.pdf [12] Plantix, \"AI-based Crop Diagnosis App,\" [Online]. Available: https://plantix.net/en/ [13] A. M. Prasad and V. Arunachalam, \"Crop yield prediction using machine learning algorithms: A case study,\" International Journal of Advanced Computer Science and Applications (IJACSA), vol. 11, no. 3, pp. 360–367, 2020. Available: https://doi.org/10.14569/IJACSA.2020.0110347 [14] M. V. Ramesh, \"Real-time smart crop protection system,\" in Proc. 6th Int. Conf. on Cloud System and Big Data Engineering, 2016, pp. 294–299. [Online]. Available: https://doi.org/10.1109/CONFLUENCE.2016.7508121 [15] V. Singh and A. K. Misra, \"Detection of plant leaf diseases using image segmentation and soft computing techniques,\" Information Processing in Agriculture, vol. 4, no. 1, pp. 41–49, 2017. [Online]. Available: https://doi.org/10.1016/j.inpa.2016.10.005

Copyright

Copyright © 2026 Disha Patel, Varsha Ahirwar, Dr. Divakar Singh, Dr. Kavita Chourasia, Dr. Kamini Maheshwar. 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.