Automated Plant Disease Classification And Prediction Using Convolutional Neural Classifier

Abstract

Hu Agriculture crops are the backbone of India's economy, which makes it one of the world's emerging nations. Crop diseases are a huge danger to food security and detecting crop diseases is still a difficult process to do. Many traditional classification techniques, such as Artificial Neural Networks (ANN), Support Vector Machines (SVM), K-nearest neighbours (k-NN), and random forests, are available for use in this context. As a result, the accuracy of such systems has saturated because they are reliant on a hand-crafted feature extraction procedure. It is necessary to add Convolutional Neural Networks (CNNs) for classification of tomato leaf disease in order to increase the accuracy of our classification method. The architecture consists of five convolution layers with 64 filters each, each with a kernel size of 3x3, a pooling size of 2x2, and a fully linked layer 2046 at the end of convolution and pooling. 3,300 Tomato crop leaf images were obtained from the Village Net dataset (healthy leaf, late blight, and yellow virus) for the purpose of assessing the performance of the crop in the studies. When it comes to crop disease classification, the experimental findings demonstrate that the suggested model achieves an accuracy of 98.18 percent.