Hybrid CNN and Transformer-Based Sequential Learning Techniques for Plant Disease Classification
Plant diseases have important consequences for livelihoods and economies, both on local and global scales, whereby the spread of plant pathogens can lead to high levels of damage to agricultural productivity. In this regard, deep learning (DL) has evolved as a promising remedy. Nevertheless, the lev...
Saved in:
| Main Authors: | , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
IEEE
2025-01-01
|
| Series: | IEEE Access |
| Subjects: | |
| Online Access: | https://ieeexplore.ieee.org/document/11072169/ |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1849714730311090176 |
|---|---|
| author | Anuruk Prommakhot Jakkrit Onshaunjit Wichian Ooppakaew Grianggai Samseemoung Jakkree Srinonchat |
| author_facet | Anuruk Prommakhot Jakkrit Onshaunjit Wichian Ooppakaew Grianggai Samseemoung Jakkree Srinonchat |
| author_sort | Anuruk Prommakhot |
| collection | DOAJ |
| description | Plant diseases have important consequences for livelihoods and economies, both on local and global scales, whereby the spread of plant pathogens can lead to high levels of damage to agricultural productivity. In this regard, deep learning (DL) has evolved as a promising remedy. Nevertheless, the level of diversity among plant species still presents a constant challenge to effective plant disease classification. This research proposes a two-stream convolution operator that combines bidirectional long short-term memory (BiLSTM) for effective feature mapping and learning. Moreover, a transformer network (TransNet) is constructed based on sequential learning techniques (SLT) using long short-term memory (LSTM), bidirectional LSTM (BiLSTM), sequence-to-sequence, and gated recurrent unit models to highlight important plant disease features. The network is trained on the PlantVillage dataset, which includes 38 species of plant diseases. The experimental results indicate that the proposed model achieves scores of 97.88%, 97.93%, and 97.62% for Accuracy, Precision, and Recall, respectively, along with a low training loss of 0.0696 and a minimal training time of 83.17 minutes. The proposed approach represents a significant improvement over previous models and demonstrates the potential for further enhancing the efficiency and accuracy of plant disease classification, thereby contributing to the advancement of modern technology in the agricultural industry. |
| format | Article |
| id | doaj-art-ea9ca363dbfc46928a2998193f99ffda |
| institution | DOAJ |
| issn | 2169-3536 |
| language | English |
| publishDate | 2025-01-01 |
| publisher | IEEE |
| record_format | Article |
| series | IEEE Access |
| spelling | doaj-art-ea9ca363dbfc46928a2998193f99ffda2025-08-20T03:13:37ZengIEEEIEEE Access2169-35362025-01-011312287612288710.1109/ACCESS.2025.358628511072169Hybrid CNN and Transformer-Based Sequential Learning Techniques for Plant Disease ClassificationAnuruk Prommakhot0https://orcid.org/0000-0003-1465-047XJakkrit Onshaunjit1https://orcid.org/0009-0001-8306-5664Wichian Ooppakaew2https://orcid.org/0009-0005-3251-4369Grianggai Samseemoung3https://orcid.org/0009-0006-6960-631XJakkree Srinonchat4https://orcid.org/0000-0002-0262-72911Department of Electronics and Telecommunication Engineering, Faculty of Engineering, Rajamangala University of Technology Thanyaburi, Khlong Hok, Khlong Luang, Pathum Thani, Thailand1Department of Electronics and Telecommunication Engineering, Faculty of Engineering, Rajamangala University of Technology Thanyaburi, Khlong Hok, Khlong Luang, Pathum Thani, Thailand1Department of Electronics and Telecommunication Engineering, Faculty of Engineering, Rajamangala University of Technology Thanyaburi, Khlong Hok, Khlong Luang, Pathum Thani, Thailand2Department of Agricultural Engineering, Faculty of Engineering, Rajamangala University of Technology Thanyaburi, Khlong Hok, Khlong Luang, Pathum Thani, Thailand1Department of Electronics and Telecommunication Engineering, Faculty of Engineering, Rajamangala University of Technology Thanyaburi, Khlong Hok, Khlong Luang, Pathum Thani, ThailandPlant diseases have important consequences for livelihoods and economies, both on local and global scales, whereby the spread of plant pathogens can lead to high levels of damage to agricultural productivity. In this regard, deep learning (DL) has evolved as a promising remedy. Nevertheless, the level of diversity among plant species still presents a constant challenge to effective plant disease classification. This research proposes a two-stream convolution operator that combines bidirectional long short-term memory (BiLSTM) for effective feature mapping and learning. Moreover, a transformer network (TransNet) is constructed based on sequential learning techniques (SLT) using long short-term memory (LSTM), bidirectional LSTM (BiLSTM), sequence-to-sequence, and gated recurrent unit models to highlight important plant disease features. The network is trained on the PlantVillage dataset, which includes 38 species of plant diseases. The experimental results indicate that the proposed model achieves scores of 97.88%, 97.93%, and 97.62% for Accuracy, Precision, and Recall, respectively, along with a low training loss of 0.0696 and a minimal training time of 83.17 minutes. The proposed approach represents a significant improvement over previous models and demonstrates the potential for further enhancing the efficiency and accuracy of plant disease classification, thereby contributing to the advancement of modern technology in the agricultural industry.https://ieeexplore.ieee.org/document/11072169/Two-stream networktransformer networkconvolutional neural networkssequential learning techniquesplant disease classification |
| spellingShingle | Anuruk Prommakhot Jakkrit Onshaunjit Wichian Ooppakaew Grianggai Samseemoung Jakkree Srinonchat Hybrid CNN and Transformer-Based Sequential Learning Techniques for Plant Disease Classification IEEE Access Two-stream network transformer network convolutional neural networks sequential learning techniques plant disease classification |
| title | Hybrid CNN and Transformer-Based Sequential Learning Techniques for Plant Disease Classification |
| title_full | Hybrid CNN and Transformer-Based Sequential Learning Techniques for Plant Disease Classification |
| title_fullStr | Hybrid CNN and Transformer-Based Sequential Learning Techniques for Plant Disease Classification |
| title_full_unstemmed | Hybrid CNN and Transformer-Based Sequential Learning Techniques for Plant Disease Classification |
| title_short | Hybrid CNN and Transformer-Based Sequential Learning Techniques for Plant Disease Classification |
| title_sort | hybrid cnn and transformer based sequential learning techniques for plant disease classification |
| topic | Two-stream network transformer network convolutional neural networks sequential learning techniques plant disease classification |
| url | https://ieeexplore.ieee.org/document/11072169/ |
| work_keys_str_mv | AT anurukprommakhot hybridcnnandtransformerbasedsequentiallearningtechniquesforplantdiseaseclassification AT jakkritonshaunjit hybridcnnandtransformerbasedsequentiallearningtechniquesforplantdiseaseclassification AT wichianooppakaew hybridcnnandtransformerbasedsequentiallearningtechniquesforplantdiseaseclassification AT grianggaisamseemoung hybridcnnandtransformerbasedsequentiallearningtechniquesforplantdiseaseclassification AT jakkreesrinonchat hybridcnnandtransformerbasedsequentiallearningtechniquesforplantdiseaseclassification |