Apple and Grape Leaf Disease Classification using MLP and CNN

Abstract

Plant leaf diseases are a common thing that every farmer experience. It also affects the production of crops resulting in significant losses to the farmers and also there will be a reduction in efficiency in the production of agricultural crops. Various crops like Apple, Grapes, Tomato, Potato, Pepper and many others get affected by the leaf diseases. Especially Apple’s leaf and Grape’s leaf gets affected by various diseases like Apple Scab, Black rot, Cedar apple Rust, Grape leaf blight, and Esca. In order to classify the leaf diseases of apple and grapes, dataset containing leaf disease images of both were obtained from the Plant Village dataset in Kaggle. The dataset had around 3171 images of apple leaf disease and 4062 images of grape’s. Algorithms like Multi Layer Perceptron (MLP) and Convolutional Neural Networks (CNN) are used for the classification of leaf diseases. The dataset is pre-processed and fed into the classifiers and classification of the leaf diseases is performed. The performance measures and the accuracy of the classifiers are compared and analysed.