Disease Detection in Cotton Plants Using Deep Learning

EasyChair Preprint 11525

Abstract

This article suggests utilizing deep learning models
to classify cotton leaves from images captured on the field as
a means of identifying any potential lessons. The scourge of
agricultural pests and diseases looms large, especially in tropical
regions where cotton cultivation is widespread. The pernicious
menace has the potential to severely impede crop yields and
inflict major financial losses on farmers. Effective solutions
are needed for these problems; however, initial symptoms can
be challenging to differentiate between making it difficult for
farmers to correctly identify lesions. To address this issue,
researchers have proposed using deep learning methods that
allow monitoring of crop health and better management decisionmaking
through screening of cotton leaves. The use of automatic
classifier CNN will assist with classification based on training
samples gathered from two categories resulting in low error rates
during training and improved accuracy when classifying new
data examined by our simulation results thus far suggest success
within implemented networks at minimum overall detriment or
deviation among other variations tested so far respectively.

Keyphrases: CNN, DeepLearning, InceptionV3, ResNet50, VGG-16

@booklet{EasyChair:11525,
  author    = {G Valarmathy and N Umapathi and S S Komagal and R Nivedha and S Rajeswari},
  title     = {Disease Detection in Cotton Plants Using Deep Learning},
  howpublished = {EasyChair Preprint 11525},
  year      = {EasyChair, 2023}}