A Novel Approach for Tomato Leaf Disease Classification with Deep Convolutional Neural Networks
Computer-aided automation systems for the detection of plant diseases represent a challenging and highly impactful research domain in the field of agriculture. Tomatoes, a major and globally significant agricultural commodity, are cultivated in large quantities. This study introduces a novel approac...
Saved in:
| Main Authors: | , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Ankara University
2024-03-01
|
| Series: | Journal of Agricultural Sciences |
| Subjects: | |
| Online Access: | https://dergipark.org.tr/tr/download/article-file/3286750 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Summary: | Computer-aided automation systems for the detection of plant diseases represent a challenging and highly impactful research domain in the field of agriculture. Tomatoes, a major and globally significant agricultural commodity, are cultivated in large quantities. This study introduces a novel approach for the automated detection of diseases on tomato leaves, leveraging both classical machine learning methods and deep neural networks for image classification. Specifically, classical learning methods employed the local binary pattern (LBP) technique for feature extraction, while classification tasks were carried out using extreme learning machines, k-nearest neighborhood (kNN), and support vector machines (SVM). In contrast, a novel convolutional neural network (CNN) framework, complete with unique parameters and layers, was utilized for deep learning. The results of this study demonstrate that the proposed approach outperforms state-of-the-art studies in terms of accuracy. The classification process covered various scenarios, including binary classification (healthy vs. unhealthy), 6-class classification, and 10-class classification for distinguishing different types of diseases. The findings indicate that the CNN model consistently outperformed classical learning methods, achieving accuracy rates of 99.5%, 98.50%, and 97.0% for 2-class, 6-class, and 10-class classifications, respectively. Future research may explore the use of computer-aided automated systems to detect diseases in diverse plant species. |
|---|---|
| ISSN: | 1300-7580 2148-9297 |