A Convolutional Neural Network Model for Prediction of Paddy Crop Disease

Abstract

Rice is a dietary staple all through Indian Subcontinent, notably in Bangladesh. The staple food consumed by 130 million people in Bangladesh is rice. To feed its 130 million people, Bangladesh currently produces 25 million tons of rice. If massive amounts of rice weren't thrown away every year due to disorganization, the number may be higher. Bangladesh is an evolving nation, hence the dropout index hasn't dropped slightly, whereas. The majority of cultivators are ignorant of the various types of rice diseases. Therefore, it is genuinely intriguing research to learn more about the condition using the resource of afflicted needles in farming activities. This study report consists of a pilot for the identification and classification of Paddy sickness utilizing inflamed leaves and algorithmic learning devices. We must not overlook the three rice ailments known as Blast disease, Plant Hopper ailment, and Leaf Folder condition. Right here, we've adopted a category-based deep learning version rooted in CNN. Notably, we utilized a diversity of switch mastering versions for classification. The meticulous pre-Mannering of the photograph is the most crucial aspect of this study. After conducting numerical inference, we mentored our strategy and evaluated it utilizing the dataset. In the end, we examined other approaches, but CNN produced handy findings for our dataset, with an exactness of roughly 99.89%.