A Hybrid Approach Using Convolutional Neural Network Model and Image Processing for Crop Disease Detection

Plant and crop cultivation rates are steadily growing over the world as human and animal demands rise. Plant disease, on the other hand, is a persistent problem for smallholder farmers, jeopardizing their livelihoods and food security. Using technologies like image processing and deep learning, we can successfully detect plant diseases in their early stages. The entire process of putting this ailment diagnosis model into practice is described in detail throughout the paper, beginning with the collection of images to create a database. Deep learning frameworks (such as convolutional neural networks (CNNs)) have made significant progress in image processing fine-tuning to match a database of a plant's leaves generated independently for different plant diseases. The web application for the developed model, which can recognize plant illnesses, is now available. A collection of leaf photographs acquired in a controlled situation is used to train and evaluate the model. Validation data shows that the suggested technique is 86 percent accurate.