From laboratory to field: cross-domain few-shot learning for crop disease identification in the field
Few-shot learning (FSL) methods have made remarkable progress in the field of plant disease recognition, especially in scenarios with limited available samples. However, current FSL approaches are usually limited to a restrictive setting where base classes and novel classes come from the same domain...
Saved in:
| Main Authors: | , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Frontiers Media S.A.
2024-12-01
|
| Series: | Frontiers in Plant Science |
| Subjects: | |
| Online Access: | https://www.frontiersin.org/articles/10.3389/fpls.2024.1434222/full |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1846118152407089152 |
|---|---|
| author | Sen Yang Quan Feng Jianhua Zhang Wanxia Yang Wenwei Zhou Wenbo Yan |
| author_facet | Sen Yang Quan Feng Jianhua Zhang Wanxia Yang Wenwei Zhou Wenbo Yan |
| author_sort | Sen Yang |
| collection | DOAJ |
| description | Few-shot learning (FSL) methods have made remarkable progress in the field of plant disease recognition, especially in scenarios with limited available samples. However, current FSL approaches are usually limited to a restrictive setting where base classes and novel classes come from the same domain such as PlantVillage. Consequently, when the model is generalized to new domains (field disease datasets), its performance drops sharply. In this work, we revisit the cross-domain performance of existing FSL methods from both data and model perspectives, aiming to better achieve cross-domain generalization of disease by exploring inter-domain correlations. Specifically, we propose a broader cross-domain few-shot learning(CD-FSL) framework for crop disease identification that allows the classifier to generalize to previously unseen categories and domains. Within this framework, three representative CD-FSL models were developed by integrating the Brownian distance covariance (BCD) module and improving the general feature extractor, namely metric-based CD-FSL(CDFSL-BDC), optimization-based CD-FSL(CDFSL-MAML), and non-meta-learning-based CD-FSL (CDFSL-NML). To capture the impact of domain shift on model performance, six public datasets with inconsistent feature distributions between domains were selected as source domains. We provide a unified testbed to conduct extensive meta-training and meta-testing experiments on the proposed benchmarks to evaluate the generalization performance of CD-FSL in the disease domain. The results showed that the accuracy of the three CD-FSL models improved significantly as the inter-domain similarity increased. Compared with other state-of-the-art CD-FSL models, the CDFSL-BDC models had the best average performance under different domain gaps. Shifting from the pest domain to the crop disease domain, the CDFSL-BDC model achieved an accuracy of 63.95% and 80.13% in the 1-shot/5-shot setting, respectively. Furthermore, extensive evaluation on a multi-domain datasets demonstrated that multi-domain learning exhibits stronger domain transferability compared to single-domain learning when there is a large domain gap between the source and target domain. These comparative results suggest that optimizing the CD-FSL method from a data perspective is highly effective for solving disease identification tasks in field environments. This study holds promise for expanding the application of deep learning techniques in disease detection and provides a technical reference for cross-domain disease detection. |
| format | Article |
| id | doaj-art-caa1f6d4b3e74c4595138b00c5b666e7 |
| institution | Kabale University |
| issn | 1664-462X |
| language | English |
| publishDate | 2024-12-01 |
| publisher | Frontiers Media S.A. |
| record_format | Article |
| series | Frontiers in Plant Science |
| spelling | doaj-art-caa1f6d4b3e74c4595138b00c5b666e72024-12-18T06:43:39ZengFrontiers Media S.A.Frontiers in Plant Science1664-462X2024-12-011510.3389/fpls.2024.14342221434222From laboratory to field: cross-domain few-shot learning for crop disease identification in the fieldSen Yang0Quan Feng1Jianhua Zhang2Wanxia Yang3Wenwei Zhou4Wenbo Yan5College of Mechanical and Electrical Engineering, Gansu Agricultural University, Lanzhou, ChinaCollege of Mechanical and Electrical Engineering, Gansu Agricultural University, Lanzhou, ChinaAgricultural Information Institute, Chinese Academy of Agricultural Sciences, Beijing, ChinaCollege of Mechanical and Electrical Engineering, Gansu Agricultural University, Lanzhou, ChinaCollege of Mechanical and Electrical Engineering, Gansu Agricultural University, Lanzhou, ChinaCollege of Mechanical and Electrical Engineering, Gansu Agricultural University, Lanzhou, ChinaFew-shot learning (FSL) methods have made remarkable progress in the field of plant disease recognition, especially in scenarios with limited available samples. However, current FSL approaches are usually limited to a restrictive setting where base classes and novel classes come from the same domain such as PlantVillage. Consequently, when the model is generalized to new domains (field disease datasets), its performance drops sharply. In this work, we revisit the cross-domain performance of existing FSL methods from both data and model perspectives, aiming to better achieve cross-domain generalization of disease by exploring inter-domain correlations. Specifically, we propose a broader cross-domain few-shot learning(CD-FSL) framework for crop disease identification that allows the classifier to generalize to previously unseen categories and domains. Within this framework, three representative CD-FSL models were developed by integrating the Brownian distance covariance (BCD) module and improving the general feature extractor, namely metric-based CD-FSL(CDFSL-BDC), optimization-based CD-FSL(CDFSL-MAML), and non-meta-learning-based CD-FSL (CDFSL-NML). To capture the impact of domain shift on model performance, six public datasets with inconsistent feature distributions between domains were selected as source domains. We provide a unified testbed to conduct extensive meta-training and meta-testing experiments on the proposed benchmarks to evaluate the generalization performance of CD-FSL in the disease domain. The results showed that the accuracy of the three CD-FSL models improved significantly as the inter-domain similarity increased. Compared with other state-of-the-art CD-FSL models, the CDFSL-BDC models had the best average performance under different domain gaps. Shifting from the pest domain to the crop disease domain, the CDFSL-BDC model achieved an accuracy of 63.95% and 80.13% in the 1-shot/5-shot setting, respectively. Furthermore, extensive evaluation on a multi-domain datasets demonstrated that multi-domain learning exhibits stronger domain transferability compared to single-domain learning when there is a large domain gap between the source and target domain. These comparative results suggest that optimizing the CD-FSL method from a data perspective is highly effective for solving disease identification tasks in field environments. This study holds promise for expanding the application of deep learning techniques in disease detection and provides a technical reference for cross-domain disease detection.https://www.frontiersin.org/articles/10.3389/fpls.2024.1434222/fullcross-domainfew-shot learningcrop diseasesrecognitionmulti-domain |
| spellingShingle | Sen Yang Quan Feng Jianhua Zhang Wanxia Yang Wenwei Zhou Wenbo Yan From laboratory to field: cross-domain few-shot learning for crop disease identification in the field Frontiers in Plant Science cross-domain few-shot learning crop diseases recognition multi-domain |
| title | From laboratory to field: cross-domain few-shot learning for crop disease identification in the field |
| title_full | From laboratory to field: cross-domain few-shot learning for crop disease identification in the field |
| title_fullStr | From laboratory to field: cross-domain few-shot learning for crop disease identification in the field |
| title_full_unstemmed | From laboratory to field: cross-domain few-shot learning for crop disease identification in the field |
| title_short | From laboratory to field: cross-domain few-shot learning for crop disease identification in the field |
| title_sort | from laboratory to field cross domain few shot learning for crop disease identification in the field |
| topic | cross-domain few-shot learning crop diseases recognition multi-domain |
| url | https://www.frontiersin.org/articles/10.3389/fpls.2024.1434222/full |
| work_keys_str_mv | AT senyang fromlaboratorytofieldcrossdomainfewshotlearningforcropdiseaseidentificationinthefield AT quanfeng fromlaboratorytofieldcrossdomainfewshotlearningforcropdiseaseidentificationinthefield AT jianhuazhang fromlaboratorytofieldcrossdomainfewshotlearningforcropdiseaseidentificationinthefield AT wanxiayang fromlaboratorytofieldcrossdomainfewshotlearningforcropdiseaseidentificationinthefield AT wenweizhou fromlaboratorytofieldcrossdomainfewshotlearningforcropdiseaseidentificationinthefield AT wenboyan fromlaboratorytofieldcrossdomainfewshotlearningforcropdiseaseidentificationinthefield |