Published Paper Details:

Tomatoes are one of the most important and consumable crops in the world. Tomatoes have a different amount of pieces depending on the fertilization method. Leaf diseases are a major factor affecting the quantity and quality of crop yields. Therefore, it is important to properly diagnose and classify these disorders. Various types of diseases affect tomato production. Early identification of these diseases reduces their impact on tomato plants and improves good yields. Various innovative methods for identifying and classifying specific diseases are widely used. The motivation for this work is to enable farmers to accurately identify diseases early and inform them about these diseases. A convolutional neural network (CNN) is used to effectively define and classify tomato diseases. Using Google Colab, we performed a full experiment on a dataset containing 3000 images of 9 different diseased tomato leaves and 1 healthy leaf. I will walk you through the whole process. First, the input image is preprocessed to segment the target regions of images from the original image. Second, the images are further processed using various hyper parameters of the CNN model. Finally, CNN extracts other features such as colors, textures and edges from his images. The results show that the predictions of his proposed model are 98.49% accurate.