DefinePK

DefinePK hosts the largest index of Pakistani journals, research articles, news headlines, and videos. It also offers chapter-level book search.

Efficient Diagnosis of Bacterial Leaf Spot in Tomato Plants using Deep Learning CNN Models


Article Information

Title: Efficient Diagnosis of Bacterial Leaf Spot in Tomato Plants using Deep Learning CNN Models

Authors: Zohaib Ahmad, Mohammed Abdulaziz Alfehaid, Saira Asghar, Abdul Manan, Sidra Gill, Memoona Bibi

Journal: Plant Bulletin

HEC Recognition History
Category From To
Y 2024-10-01 2025-12-31
Y 2023-07-01 2024-09-30

Publisher: Airanam Research

Country: Pakistan

Year: 2023

Volume: 2

Issue: 2

Language: English

DOI: 10.55627/pbulletin.002.02.0426

Keywords: Bacterial spotCNN ModelTomato DiseaseXantho Net

Categories

Abstract

Latest field surveys show that the bacterial spot caused by Xanthomonas campestris pv. vesicatoria (Xcv) has devastated commercial tomato output all over the world. The purpose of this study is to more quickly and accurately diagnose bacterial leaf spot problems in tomato plants. To do this, a deep learning-based Convolutional Neural Network (CNN) named Xantho_Net has been proposed to classify the bacterial and non-bacterial tomato leafs. Various state-of-art pre-trained models i.e., VGG16, MobileNet, DenseNet_201, Inception-ResNet_v2 were selected for comparison purposes. Experiments proved that the proposed CNN model is more accurate than other models with 99% accuracy.


Research Objective

To develop a deep learning-based Convolutional Neural Network (CNN) model, named Xantho_Net, for the rapid and accurate diagnosis of bacterial leaf spot in tomato plants.


Methodology

The study utilized a dataset of 2,000 tomato leaf images (healthy and infected with bacterial spot) from Kaggle. The proposed Xantho_Net model was developed and compared against state-of-the-art pre-trained CNN models: VGG16, MobileNet, DenseNet_201, and Inception-ResNet_v2. The dataset was split into 70% for training, 15% for validation, and 15% for testing. Performance was evaluated using accuracy, precision, recall, and F1-score.

Methodology Flowchart
                        graph TD
    A[Data Collection: 2000 Tomato Leaf Images] --> B[Data Preprocessing & Splitting: Train/Validation/Test];
    B --> C[Model Development: Xantho_Net];
    C --> D[Comparative Analysis with VGG16, MobileNet, DenseNet_201, Inception-ResNet_v2];
    D --> E[Model Training & Evaluation];
    E --> F[Performance Metrics: Accuracy, Precision, Recall, F1-Score];
    F --> G[Results & Conclusion];                    

Discussion

The study highlights the effectiveness of deep learning, specifically CNNs, for accurate and efficient plant disease diagnosis. The Xantho_Net model's performance indicates its potential to assist farmers in timely disease management, leading to improved crop yield and quality. The model's efficiency in terms of speed and parameter count is also noted as an advantage.


Key Findings

The proposed Xantho_Net model achieved 100% training accuracy and 99% testing accuracy. It outperformed the compared pre-trained models, demonstrating superior efficiency and accuracy in classifying bacterial and non-bacterial tomato leaves.


Conclusion

The Xantho_Net model provides an effective solution for the early and accurate diagnosis of bacterial leaf spot in tomato plants. Its high accuracy and efficiency suggest it can be a valuable tool for agricultural applications, enabling farmers to mitigate crop losses and enhance productivity.


Fact Check

* The study used 2,000 images for analysis.
* The proposed Xantho_Net model achieved 99% accuracy on the test set.
* The research was published in Plant Bulletin Vol 2, (2) 2023.


Mind Map

Loading PDF...

Loading Statistics...