AI-Driven Plant Leaf Disease Detection for Modern Agriculture

Abstract

Due to the diseases that affect the crop, farmers as well as the buyers face a critical loss. About 60% of the farmers confront losses in crop yield. As a result, there have been numerous reports of deaths of the farmers. Later progressions in artificial intelligence and through the use of deep learning techniques, automated systems are distinguished and also recognize infections in images. This model can extract the features of the disease that's shown within the given image. In this literature survey the authors recognized the tomato crop diseases and focused on certain aspects which include image dataset, no. of diseases (classes), precision of the model etc. They created a model using convolution neural network (CNN) for classifying images and explainable artificial intelligence (AI) by using a local interpretability technique called as local interpretable model-agnostic explanations (LIME) to explain the predictions that are made by the model. Evaluation of the images from the tomato disease image dataset shows that our model's accuracy is 97.78%.