Machine Learning Approach for Detection of Diseased Rice Leaves Using a Convolutional Neural Network Architecture

Read full text Summarise with VIRAT
Authors: Subrata Sinha, SAURAV MALI, Mridul Bhuyan, Amit Kumar Pathak
Cited by: 0
References: 23

Summary

No summary yet. VIRAT can read this paper and write a plain-language summary of its aim, methods, key findings and limitations.

Summarise with VIRAT Free to try with your Science Catalogs account.

Abstract

Rice (Oryza sativa) stands as a major staple crop extensively consumed not only in India but also globally. However, its production is affected by many biotic and abiotic factors. Infestations by various bacterial and viral pests is responsible for hampering the overall production of rice crops. Early detection of the diseased plants is a necessity to increase the overall production and improve the quality of rice crops. Hence automatic identification of diseased rice crop can assist the farmers for better care of the crop. The application of a deep learning approach utilizing the Convolutional Neural Network (CNN) algorithm has proven highly effective in the detection of diseases from images of rice leaves. In this approach, a model has been developed, employing a Convolutional Neural Network architecture, to achieve accurate detection of diseased rice plants. The model was trained on a dataset of 2993 images, divided into 6 classes, out of which 5 classes are of various rice leaf diseases. It achieved an accuracy of 96%.

Fields of Study & Subjects

Smart Agriculture and AI Advanced Neural Network Applications Industrial Vision Systems and Defect Detection

Publication details

Journal / source
Environment and Ecology
Volume
44
Issue
1
Pages
166–181
Publication date
1 March 2026
Language
EN