Sen Yang, Quan Feng, Jianhua Zhang, Wanxia Yang, Wenwei Zhou, Wenbo Yan
{"title":"From laboratory to field: cross-domain few-shot learning for crop disease identification in the field.","authors":"Sen Yang, Quan Feng, Jianhua Zhang, Wanxia Yang, Wenwei Zhou, Wenbo Yan","doi":"10.3389/fpls.2024.1434222","DOIUrl":null,"url":null,"abstract":"<p><p>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.</p>","PeriodicalId":12632,"journal":{"name":"Frontiers in Plant Science","volume":"15 ","pages":"1434222"},"PeriodicalIF":4.1000,"publicationDate":"2024-12-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC11688362/pdf/","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Frontiers in Plant Science","FirstCategoryId":"99","ListUrlMain":"https://doi.org/10.3389/fpls.2024.1434222","RegionNum":2,"RegionCategory":"生物学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"2024/1/1 0:00:00","PubModel":"eCollection","JCR":"Q1","JCRName":"PLANT SCIENCES","Score":null,"Total":0}
引用次数: 0
Abstract
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.
期刊介绍:
In an ever changing world, plant science is of the utmost importance for securing the future well-being of humankind. Plants provide oxygen, food, feed, fibers, and building materials. In addition, they are a diverse source of industrial and pharmaceutical chemicals. Plants are centrally important to the health of ecosystems, and their understanding is critical for learning how to manage and maintain a sustainable biosphere. Plant science is extremely interdisciplinary, reaching from agricultural science to paleobotany, and molecular physiology to ecology. It uses the latest developments in computer science, optics, molecular biology and genomics to address challenges in model systems, agricultural crops, and ecosystems. Plant science research inquires into the form, function, development, diversity, reproduction, evolution and uses of both higher and lower plants and their interactions with other organisms throughout the biosphere. Frontiers in Plant Science welcomes outstanding contributions in any field of plant science from basic to applied research, from organismal to molecular studies, from single plant analysis to studies of populations and whole ecosystems, and from molecular to biophysical to computational approaches.
Frontiers in Plant Science publishes articles on the most outstanding discoveries across a wide research spectrum of Plant Science. The mission of Frontiers in Plant Science is to bring all relevant Plant Science areas together on a single platform.