Automated Diagnosis of Rice Plant Leaves Diseases with Convolutional Neural Network

Main Article Content

Preeti Yadav, Parvinder Singh

Abstract

Rice cultivation is a cornerstone of India's agricultural sector, producing an estimated 120 million tones annually. However, this vital crop is threatened by various fungal pathogens, leading to substantial financial setbacks for growers. Key afflictions such as bacterial leaf blight, brown spot, and leaf smut not only diminish yield but also cause significant monetary losses. Conventionally, disease detection in rice relies on manual scrutiny for symptomatic evidence, a method that is both time-intensive and impractical for large-scale farming operations. Addressing this inefficiency, the present research introduces a deep learning-based approach that facilitates early detection of plant diseases through image analysis, allowing for timely and effective treatment. Leveraging cutting-edge deep learning algorithms, the research showcases the ability to accurately diagnose and address rice plant diseases in their nascent stages, with the convolutional neural network architecture, particularly ResNet101, demonstrating a high accuracy rate of 95.83% in multiclass detection and 98.65% in binary classification. This innovative strategy represents a transformative advancement in agricultural practices, providing a robust tool for enhancing rice disease management.

Article Details

Section
Articles