Paper Key : IRJ************115
Author: Pratik Palkar
Date Published: 17 Nov 2024
Abstract
: Sugarcane is a major crop worldwide and the main source of both sugar and ethanol. India's economy is mostly based on agriculture, which is the country's main source of income. Sugarcane is widely farmed in several Indian states, including Bihar, Karnataka, Tamil Nadu, Uttar Pradesh, Maharashtra, and many more, because it is a crop that grows again. The proliferation of diseases that infect sugarcane, forcing farmers to eradicate disease-ridden areas and costing small-scale farmers money if the infections are not treated and detected early, is one issue specifically impacting the sugar business. In agricultural settings, viruses and bacteria are secondary sources of plant infection, while natural disasters like storms and floods are the primary sources of damage to crops. The decline in quality might also be attributed to viral illnesses. Disease control is crucial for maintaining crop quality. When specialized agricultural specialists assist farmers with vast farming experience, they can usually diagnose ailments in crops. Occasionally, variations in weather might make it difficult to diagnose certain illnesses. It restricts the ability to diagnose problems of sugarcane. In order to address this problem, we suggested machine learning via computer vision using deep learning approaches in this study. This work trained and evaluated a deep learning model with 13,842 sugarcane photo datasets that showed both disease-infected and healthy leaves, with an accuracy of 95%. The trained model achieved this by recognizing and classifying sugarcane pictures into groups that represented healthy, sick, or diseased sugarcane leaves. As a result, this study provides a deep learning algorithm-based approach for classifying and diagnosing sugarcane diseases to support farmers.
DOI Requested