Deep-Learning-Based Classification of Bangladeshi Medicinal Plants Using Neural Ensemble Models
This research addresses the lack of publicly available datasets for Bangladeshi medicinal plants by presenting a comprehensive dataset comprising 5000 images of ten species collected under controlled conditions. To improve performance, several preprocessing techniques were employed, such as image se...
| Main Authors: | , , , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2023-08-01
|
| Series: | Mathematics |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2227-7390/11/16/3504 |
| _version_ | 1797583965599563776 |
|---|---|
| author | A. Hasib Uddin Yen-Lin Chen Bijly Borkatullah Mst. Sathi Khatun Jannatul Ferdous Prince Mahmud Jing Yang Chin Soon Ku Lip Yee Por |
| author_facet | A. Hasib Uddin Yen-Lin Chen Bijly Borkatullah Mst. Sathi Khatun Jannatul Ferdous Prince Mahmud Jing Yang Chin Soon Ku Lip Yee Por |
| author_sort | A. Hasib Uddin |
| collection | DOAJ |
| description | This research addresses the lack of publicly available datasets for Bangladeshi medicinal plants by presenting a comprehensive dataset comprising 5000 images of ten species collected under controlled conditions. To improve performance, several preprocessing techniques were employed, such as image selection, background removal, unsharp masking, contrast-limited adaptive histogram equalization, and morphological gradient. Then, we applied five state-of-the-art deep learning models to achieve benchmark performance on the dataset: VGG16, ResNet50, DenseNet201, InceptionV3, and Xception. Among these models, DenseNet201 demonstrated the highest accuracy of 85.28%. In addition to benchmarking the deep learning models, three novel neural network architectures were developed: dense-residual–dense (DRD), dense-residual–ConvLSTM-dense (DRCD), and inception-residual–ConvLSTM-dense (IRCD). The DRCD model achieved the highest accuracy of 97%, surpassing the benchmark performances of individual models. This highlights the effectiveness of the proposed architectures in capturing complex patterns and dependencies within the data. To further enhance classification accuracy, an ensemble approach was adopted, employing both hard ensemble and soft ensemble techniques. The hard ensemble achieved an accuracy of 98%, while the soft ensemble achieved the highest accuracy of 99%. These results demonstrate the effectiveness of ensembling techniques in boosting overall classification performance. The outcomes of this study have significant implications for the accurate identification and classification of Bangladeshi medicinal plants. This research provides valuable resources for traditional medicine, drug discovery, and biodiversity conservation efforts. The developed models and ensemble techniques can aid researchers, botanists, and practitioners in accurately identifying medicinal plant species, thereby facilitating the utilization of their therapeutic potential and contributing to the preservation of biodiversity. |
| first_indexed | 2024-03-10T23:45:37Z |
| format | Article |
| id | doaj.art-720c3dc1823348b0af8dd8c6fe871afd |
| institution | Directory Open Access Journal |
| issn | 2227-7390 |
| language | English |
| last_indexed | 2024-03-10T23:45:37Z |
| publishDate | 2023-08-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Mathematics |
| spelling | doaj.art-720c3dc1823348b0af8dd8c6fe871afd2023-11-19T02:02:58ZengMDPI AGMathematics2227-73902023-08-011116350410.3390/math11163504Deep-Learning-Based Classification of Bangladeshi Medicinal Plants Using Neural Ensemble ModelsA. Hasib Uddin0Yen-Lin Chen1Bijly Borkatullah2Mst. Sathi Khatun3Jannatul Ferdous4Prince Mahmud5Jing Yang6Chin Soon Ku7Lip Yee Por8Department of Computer Science and Engineering, Khwaja Yunus Ali University, Sirajganj 6751, BangladeshDepartment of Computer Science and Information Engineering, National Taipei University of Technology, Taipei 106344, TaiwanDepartment of Computer Science and Engineering, Khwaja Yunus Ali University, Sirajganj 6751, BangladeshDepartment of Computer Science and Engineering, Khwaja Yunus Ali University, Sirajganj 6751, BangladeshDepartment of Computer Science and Engineering, Jannat Ara Henry Science & Technology College, Sirajganj 6700, BangladeshDepartment of Computer Science and Engineering, Chandpur Science and Technology University, Chandpur 3600, BangladeshDepartment of Computer System and Technology, Faculty of Computer Science and Information Technology, Universiti Malaya, Kuala Lumpur 50603, MalaysiaDepartment of Computer Science, Universiti Tunku Abdul Rahman, Kampar 31900, MalaysiaDepartment of Computer System and Technology, Faculty of Computer Science and Information Technology, Universiti Malaya, Kuala Lumpur 50603, MalaysiaThis research addresses the lack of publicly available datasets for Bangladeshi medicinal plants by presenting a comprehensive dataset comprising 5000 images of ten species collected under controlled conditions. To improve performance, several preprocessing techniques were employed, such as image selection, background removal, unsharp masking, contrast-limited adaptive histogram equalization, and morphological gradient. Then, we applied five state-of-the-art deep learning models to achieve benchmark performance on the dataset: VGG16, ResNet50, DenseNet201, InceptionV3, and Xception. Among these models, DenseNet201 demonstrated the highest accuracy of 85.28%. In addition to benchmarking the deep learning models, three novel neural network architectures were developed: dense-residual–dense (DRD), dense-residual–ConvLSTM-dense (DRCD), and inception-residual–ConvLSTM-dense (IRCD). The DRCD model achieved the highest accuracy of 97%, surpassing the benchmark performances of individual models. This highlights the effectiveness of the proposed architectures in capturing complex patterns and dependencies within the data. To further enhance classification accuracy, an ensemble approach was adopted, employing both hard ensemble and soft ensemble techniques. The hard ensemble achieved an accuracy of 98%, while the soft ensemble achieved the highest accuracy of 99%. These results demonstrate the effectiveness of ensembling techniques in boosting overall classification performance. The outcomes of this study have significant implications for the accurate identification and classification of Bangladeshi medicinal plants. This research provides valuable resources for traditional medicine, drug discovery, and biodiversity conservation efforts. The developed models and ensemble techniques can aid researchers, botanists, and practitioners in accurately identifying medicinal plant species, thereby facilitating the utilization of their therapeutic potential and contributing to the preservation of biodiversity.https://www.mdpi.com/2227-7390/11/16/3504Bangladeshi medicinal plantsmedicinal plantdeep learningclassificationneural ensemble methods |
| spellingShingle | A. Hasib Uddin Yen-Lin Chen Bijly Borkatullah Mst. Sathi Khatun Jannatul Ferdous Prince Mahmud Jing Yang Chin Soon Ku Lip Yee Por Deep-Learning-Based Classification of Bangladeshi Medicinal Plants Using Neural Ensemble Models Mathematics Bangladeshi medicinal plants medicinal plant deep learning classification neural ensemble methods |
| title | Deep-Learning-Based Classification of Bangladeshi Medicinal Plants Using Neural Ensemble Models |
| title_full | Deep-Learning-Based Classification of Bangladeshi Medicinal Plants Using Neural Ensemble Models |
| title_fullStr | Deep-Learning-Based Classification of Bangladeshi Medicinal Plants Using Neural Ensemble Models |
| title_full_unstemmed | Deep-Learning-Based Classification of Bangladeshi Medicinal Plants Using Neural Ensemble Models |
| title_short | Deep-Learning-Based Classification of Bangladeshi Medicinal Plants Using Neural Ensemble Models |
| title_sort | deep learning based classification of bangladeshi medicinal plants using neural ensemble models |
| topic | Bangladeshi medicinal plants medicinal plant deep learning classification neural ensemble methods |
| url | https://www.mdpi.com/2227-7390/11/16/3504 |
| work_keys_str_mv | AT ahasibuddin deeplearningbasedclassificationofbangladeshimedicinalplantsusingneuralensemblemodels AT yenlinchen deeplearningbasedclassificationofbangladeshimedicinalplantsusingneuralensemblemodels AT bijlyborkatullah deeplearningbasedclassificationofbangladeshimedicinalplantsusingneuralensemblemodels AT mstsathikhatun deeplearningbasedclassificationofbangladeshimedicinalplantsusingneuralensemblemodels AT jannatulferdous deeplearningbasedclassificationofbangladeshimedicinalplantsusingneuralensemblemodels AT princemahmud deeplearningbasedclassificationofbangladeshimedicinalplantsusingneuralensemblemodels AT jingyang deeplearningbasedclassificationofbangladeshimedicinalplantsusingneuralensemblemodels AT chinsoonku deeplearningbasedclassificationofbangladeshimedicinalplantsusingneuralensemblemodels AT lipyeepor deeplearningbasedclassificationofbangladeshimedicinalplantsusingneuralensemblemodels |