Published Paper Details:

Abstract: Each year, several tomato plant illnesses cause farmers to lose money and squander their crops. These diseases of the tomato plant are not easily known. Though tomatoes can be purchased, we will never know the status of the tomato plant. To overcome this and to make farmers and consumers have a happy and healthy life, we can design a web application that detects the status of the tomato plant. Our research focuses on the construction of a strong and accurate classification model capable of identifying several illnesses affecting tomato leaves, including early blight, late blight, bacterial spot, and mosaic virus. We deployed a diversified dataset comprising high-resolution photos of damaged and healthy tomato leaves, obtained from different places and under varying environmental conditions. The results illustrate the model's high accuracy, sensitivity, and specificity in differentiating between distinct tomato leaf illnesses and healthy leaves. The project seeks to develop a web application that leverages image classification using convolutional neural networks (CNN) to help tomato growers detect and prevent plant illnesses, which are a major cause of economic loss and crop waste. The program will enable farmers to upload a picture of their tomato plants and obtain an immediate evaluation of whether the plant is healthy or infected. By doing so, we can become familiar with the position of the particular tomato and the tomato plant. The technology stack for this project will contain TensorFlow for model creation, data augmentation, and the TF dataset, as well as React JS, which is used for front-end development, and Flask for back-end development.