Leaf Disease image classification method based on improved convolutional neural network

Abstract

This paper attempts to apply neural networks to classify maize disease images. The classification model is based on thousands of images of actual maize diseased leaves, which are relatively belonging to maize leaf blight leaves, corn rusty leaves, corn gray spot disease leaves, and corn healthy leaves. Classification neural networks are a popular machine learning method that will serve as an adjunct to help diagnose whether corn leaves are diseased. This article analyzes and contains details of the algorithms we use. This article demonstrates our proposed VGG-16-based neural network model. The average recognition rate of the final model is 94.64%.