Early Disease Detection in Potato Plant using Deep Learning for Precise Agriculture

Potatoes are a key staple food worldwide, ranking fourth in global consumption. The COVID-19 pandemic has further increased potato consumption, but the crop’s cultivation faces significant challenges due to diseases like Common Scab, Black Scurf, Black Leg, and Pink Rot, particularly in India. These diseases can severely reduce potato yields, leading to major financial losses for farmers. Given the crop’s economic significance, improving disease detection and management is crucial. Initial detection with accurate detection of these diseases is crucial for better crop management, which ultimately supports a sustainable food supply chain. Traditional methods for detecting potato diseases often fall short, especially for early intervention. As a result, advanced technologies like Deep Learning (DL) offer promising alternatives. This proposed system uses Convolutional Neural Networks (CNNs) to classify potato leaves as diseased or healthy, focusing on five categories: Early Blight, Healthy, Late Blight, Non-potato, and object. A simple CNN model is trained and tested for image classification tasks. By developing a web interface that allows farmers to upload leaf images, the proposed system aims to provide an easy-to-use tool for early disease identification.