An Optimized Deep Learning Model for Grassy Shoot Disease Prediction in Sugarcane
Vishesh Tanwar, Shweta Lamba, Bhanu Sharma, Avinash Sharma
Abstract
Throughout an August-September 20222 study of sugarcane fields at the Sugarcane Research Institute (SRI) in Shahjahanpur, UP, India, a 6% to 28% prevalence of sugarcane grassy shoot disease was reported in several sugarcane varieties. Traditional approaches for disease prediction in sugarcane crops include drawbacks such as high expenses for obtaining the data input required to run the algorithm, a shortage of geographic information, or an inadequate dataset. In this paper, we combine the convolutional neural networks (CNN) model for Grassy shoot sickness prediction and classification, as well as explore the sugarcane issue. This paper investigates the sugarcane grassy shoot infection. The diseased plant is imaged utilizing secondary resources first. The greatest aspects of CNN's deep learning algorithms are utilized to extract and classify the illness's attributes. The study had a 96% prediction rate for Grassy shoot Sugarcane Disease.