Coffee Plant Leaf Disease Detection Using Machine Learning

Abstract The coffee industry faces significant challenges due to the widespread occurrence of various diseases that affect the health and productivity of coffee plants. Early detection and proper management of these diseases are crucial for ensuring crop health and maximizing yield. This paper presents a web-based system for detecting coffee leaf diseases using a custom vanilla Convolutional Neural Network (CNN) model. The system is trained on a dataset of 10,000 images classified into five categories: healthy, miner, phoma, red spider mite, and rust. The model, built using TensorFlow and Keras, is integrated into a Flask web application where users can upload images of coffee leaves, and the system predicts the disease while providing the cause and cure information. The proposed approach offers an efficient solution for automated disease detection, particularly suitable for limited datasets and smaller-scale agricultural operations.