Anaugmented attention-based lightweight CNN model for plant water stress detection
Recently, deep learning techniques specifically the Convolutional Neural Networks (CNNs) have reported outstanding results from the application for plant water stress detection based on computer vision system compared to other machine learning methods. However, the size of the conventional CNN model...
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Published: |
Springer
2023
|
| _version_ | 1817927388107898880 |
|---|---|
| author | Kamarudin, Mohd Hider Ismail, Zool Hilmi Saidi, Noor Baity Hanada, Kousuke |
| author_facet | Kamarudin, Mohd Hider Ismail, Zool Hilmi Saidi, Noor Baity Hanada, Kousuke |
| author_sort | Kamarudin, Mohd Hider |
| collection | UPM |
| description | Recently, deep learning techniques specifically the Convolutional Neural Networks (CNNs) have reported outstanding results from the application for plant water stress detection based on computer vision system compared to other machine learning methods. However, the size of the conventional CNN models is generally too large for its deployment on resource-limited devices such as mobile smartphone or embedded devices. In this study, a lightweight CNN is proposed by incorporating attention mechanism as an augmentation module into the model. The model was trained, validated, and tested using plant images of Setaria grass undergone three water stress treatments. Experimental results show that the proposed method improved the interclass precision, recall, F1-score, and the overall accuracy by more than 9. Compared to the established lightweight CNN models, the proposed lightweight CNN achieved faster computational time with comparable parameters. In addition, the proposed lightweight model is also efficient when trained on small plant dataset with limited overfitting. |
| first_indexed | 2024-12-09T02:17:39Z |
| format | Article |
| id | upm.eprints-106545 |
| institution | Universiti Putra Malaysia |
| last_indexed | 2024-12-09T02:17:39Z |
| publishDate | 2023 |
| publisher | Springer |
| record_format | dspace |
| spelling | upm.eprints-1065452024-09-26T07:39:28Z http://psasir.upm.edu.my/id/eprint/106545/ Anaugmented attention-based lightweight CNN model for plant water stress detection Kamarudin, Mohd Hider Ismail, Zool Hilmi Saidi, Noor Baity Hanada, Kousuke Recently, deep learning techniques specifically the Convolutional Neural Networks (CNNs) have reported outstanding results from the application for plant water stress detection based on computer vision system compared to other machine learning methods. However, the size of the conventional CNN models is generally too large for its deployment on resource-limited devices such as mobile smartphone or embedded devices. In this study, a lightweight CNN is proposed by incorporating attention mechanism as an augmentation module into the model. The model was trained, validated, and tested using plant images of Setaria grass undergone three water stress treatments. Experimental results show that the proposed method improved the interclass precision, recall, F1-score, and the overall accuracy by more than 9. Compared to the established lightweight CNN models, the proposed lightweight CNN achieved faster computational time with comparable parameters. In addition, the proposed lightweight model is also efficient when trained on small plant dataset with limited overfitting. Springer 2023-04-24 Article PeerReviewed Kamarudin, Mohd Hider and Ismail, Zool Hilmi and Saidi, Noor Baity and Hanada, Kousuke (2023) Anaugmented attention-based lightweight CNN model for plant water stress detection. Applied Intelligence, 53. pp. 20828-20843. ISSN 0924-669X; ESSN: 1573-7497 https://link.springer.com/article/10.1007/s10489-023-04583-8 10.1007/s10489-023-04583-8 |
| spellingShingle | Kamarudin, Mohd Hider Ismail, Zool Hilmi Saidi, Noor Baity Hanada, Kousuke Anaugmented attention-based lightweight CNN model for plant water stress detection |
| title | Anaugmented attention-based lightweight CNN model for plant water stress detection |
| title_full | Anaugmented attention-based lightweight CNN model for plant water stress detection |
| title_fullStr | Anaugmented attention-based lightweight CNN model for plant water stress detection |
| title_full_unstemmed | Anaugmented attention-based lightweight CNN model for plant water stress detection |
| title_short | Anaugmented attention-based lightweight CNN model for plant water stress detection |
| title_sort | anaugmented attention based lightweight cnn model for plant water stress detection |
| work_keys_str_mv | AT kamarudinmohdhider anaugmentedattentionbasedlightweightcnnmodelforplantwaterstressdetection AT ismailzoolhilmi anaugmentedattentionbasedlightweightcnnmodelforplantwaterstressdetection AT saidinoorbaity anaugmentedattentionbasedlightweightcnnmodelforplantwaterstressdetection AT hanadakousuke anaugmentedattentionbasedlightweightcnnmodelforplantwaterstressdetection |