Design of a Machine Learning Model for Predicting Plant Disease.

Abstract

Identification of diseases from the leaf images of a tomato leaf is one of the interesting research areas within the agriculture field, that machine learning ideas of computer field can be applied. This paper presents a model for the detection and classification of tomato leaf diseases based on the images of infected tomato leaf. We consider 10 tomato diseases named Bacterial spot, Early blight, late blight, Leaf Mold, Septoria_leaf spot, Tomato mosaic virus, healthy, Spider mites Two spotted spider mite,Target Spot, Tomato Yellow Leaf Curl Virus. It can also observe Healthy leafs. During this analysis, we used deep learning based model (CNN) for classification. First, we pre-process our image dataset very carefully as a result of pre processing is that the most vital a part of our analysis. Then we train our model and validate according to the dataset. We test various techniques for this research but cnn works pretty well for our dataset, it gives an accuracy level of 87%.