AI-Driven Crop Disease Prediction and Management System

AI-Driven Crop Disease Prediction and Management System

Hima S, Varsha S, Poojitha S

CSE, Presidency University, Bengaluru, Karnataka, India

Guided By: Mr.T Ramesh, Assistant Professor, SOE, Presidency University

hima.20211cse0567@presidencyuniversity.in varsha.20211cse0576@presidencyuniversity.in poojitha.20211cse0632@presidencyuniversity.in

ABSTRACT
There are severe agricultural challenges faced by most countries such as India, because of the outbreaks of plant diseases and climate change, which jeopardize the livelihoods of millions dependent on agriculture. A web application has been designed to ameliorate this situation by giving real-time crop recommendation suggestions based on environmental parameters such as temperature, humidity, soil nutrients, pH, and rainfall. The use of AI and ML for plant disease management and crop yield prediction follows a path of transformation toward helping in accurate, data-driven decision-making for agriculture. Seven machine learning models, namely Decision Tree, Naive Bayes, Support Vector Machine (SVM), Logistic Regression, Random Forest, XGBoost, and k-Nearest Neighbors (KNN), were used to test and validate the system.

Besides, the project also incorporates a plant disease identification system based on a Convolutional Neural Network (CNN); it scans leaf images for the high-accuracy detection and classification of plant diseases.

This means farmers will be able to act early, preventing crop loss and enhancing agricultural productivity.

KEYWORDS: CNN, Crop Recommendation, Machine Learning, Plant Disease Identification, Random Forest