Enhancing Cotton Crop Health Monitoring by Deep Learning Models for Plant and Leaf Disease Classification

Abstract

The accurate classification of plant diseases is crucial for effective crop management and yield improvement. This study explores the application of deep learning models for the automated classification of cotton leaf and plant conditions. A data set comprising 2,316 images, classified into four classes: diseased cotton leaf, diseased cotton plant, fresh cotton leaf, and fresh cotton plant, was used for training and evaluation. Three well-known convolutional neural network (CNN) architectures, namely AlexNet, GoogLeNet, and SqueezeNet, were employed for classification. The experimental results indicate that GoogLeNet outperformed the other models, achieving a recognition accuracy of 73.67%, followed by SqueezeNet at 66.33% and AlexNet at 39.71 %. These findings highlight the efficiency of GoogLeNet in feature extraction and classification for cotton leaf and plant disease detection. The study demonstrates the potential of deep learning techniques in smart agriculture by automating disease detection processes.

Plant phenotyping relevance

綿花の葉・植物体の病害状態を画像から分類する深層学習手法が研究の中心であり、植物の病害表現型を直接推定しているため含める。

Code and data availability

公開本文の所在を確認できませんでした。非公開または購読が必要な可能性があります。

No evidence-backed public reproduction asset is currently recorded.