Leaf Disease Detection using Convolutional Neural Network

For the past two decades, the imbalance between food supply and population growth has been a major concern. Agriculture plays an important role in human development, and technological improvements have significantly contributed to this process. In this study, Convolutional Neural Networks (CNNs) will be utilized to identify plant leaf diseases based on leaf images. The objective is to develop an application that accurately classifies plant images as healthy or diseased. This will be achieved by collecting and preprocessing a dataset of damaged and healthy plant images under varying watering conditions. Globally, essential agricultural commodities such as tomatoes, cotton, paddy, etc. often experience price fluctuations due to supply and demand issues. Additionally, many farmers lack access to agricultural specialists for diagnosing and treating leaf diseases. To solve this issue, a low-cost image processing technique is developed to detect leaf diseases in tomato plants, an essential ingredient in Indian kitchens. Using the CNN models, farmers can compare images of diseased leaves and detect infections early, enabling timely intervention. This approach benefits both farmers and consumers by stabilizing the prices of, tomatoes as it is rapid, cost-effective, and applicable year-round.