Deep Learning-based Hybrid Model for Severity Prediction of Leaf Smut Sugarcane Infection

Abstract

Traditional models for predicting diseases in sugarcane crops show some drawbacks, including expensive costs for getting the data input needed to execute the model, a lack of spatial data, or a poor dataset. These problems are discussed in this work, which also develops a yield prediction fusion model. Convolutional neural networks (CNN) and support vector machines (SVM). make up the prediction model.In this work, the leaf smut infection of sugarcane is discussed. The sick plant is first photographed utilizing secondary sources. For feature extraction and classification of the various levels of severity of the smut infection, the best features of the deep learning techniques CNN and SVM are applied. Mild, Average, Severe, and Profound are the four seriousness prediction levels used in the study. Mendeley and Kaggle are the data repositories that were utilized, and the total size of the dataset was 950. The four severity level forecasts made by the suggested framework are 98% accurate.