VCPC:基于虚拟对比约束和原型标定的少枝类-增量植物病害分类。

IF 4.4 2区 生物学 Q1 BIOCHEMICAL RESEARCH METHODS
Lunhong Lou, Jianwu Lin, Lin You, Xin Zhang, Tomislav Cernava, Hanyu Lu, Xiaoyulong Chen
{"title":"VCPC:基于虚拟对比约束和原型标定的少枝类-增量植物病害分类。","authors":"Lunhong Lou, Jianwu Lin, Lin You, Xin Zhang, Tomislav Cernava, Hanyu Lu, Xiaoyulong Chen","doi":"10.1186/s13007-025-01423-3","DOIUrl":null,"url":null,"abstract":"<p><p>Deep learning demonstrates strong generalisation capabilities, driving substantial progress in plant disease recognition systems. However, current methods are predominantly optimised for offline implementation. Real-time crop surveillance systems encounter streaming images containing novel disease classes in few-shot conditions, demanding incrementally adaptive models. This capability is called few-shot class-incremental learning (FSCIL). Here, we introduce VCPV-virtual contrastive constraints with prototype vector calibration-enabling sustainable plant disease classification under FSClL conditions. Specifically, our method consists of two phases: the base class training phase and the incremental training phase. During the base class training phase, the virtual contrastive class constraints (VCC) module is utilised to enhance learning from base classes and allocate sufficient embedding space for new plant disease images. In the incremental training phase, the prototype calibration embedding (PCE) module is introduced to distinguish newly arriving plant disease categories from previous ones, thereby optimising the prototype space and enhancing the recognition accuracy of new categories. We evaluated our approach on the PlantVillage dataset, and the experimental results under both 5-way 5-shot and 3-way 5-shot settings demonstrate that our method achieves state-of-the-art accuracy. At the same time, we achieved promising performance on the publicly available CIFAR-100 dataset. Furthermore, the visualisation results validate that our strategy effectively supports fine-grained, sustainable disease recognition, highlighting the potential of our approach to advance FSCIL in the field of plant disease monitoring.</p>","PeriodicalId":20100,"journal":{"name":"Plant Methods","volume":"21 1","pages":"105"},"PeriodicalIF":4.4000,"publicationDate":"2025-07-31","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC12312506/pdf/","citationCount":"0","resultStr":"{\"title\":\"VCPC: virtual contrastive constraint and prototype calibration for few-shot class-incremental plant disease classification.\",\"authors\":\"Lunhong Lou, Jianwu Lin, Lin You, Xin Zhang, Tomislav Cernava, Hanyu Lu, Xiaoyulong Chen\",\"doi\":\"10.1186/s13007-025-01423-3\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"<p><p>Deep learning demonstrates strong generalisation capabilities, driving substantial progress in plant disease recognition systems. However, current methods are predominantly optimised for offline implementation. Real-time crop surveillance systems encounter streaming images containing novel disease classes in few-shot conditions, demanding incrementally adaptive models. This capability is called few-shot class-incremental learning (FSCIL). Here, we introduce VCPV-virtual contrastive constraints with prototype vector calibration-enabling sustainable plant disease classification under FSClL conditions. Specifically, our method consists of two phases: the base class training phase and the incremental training phase. During the base class training phase, the virtual contrastive class constraints (VCC) module is utilised to enhance learning from base classes and allocate sufficient embedding space for new plant disease images. In the incremental training phase, the prototype calibration embedding (PCE) module is introduced to distinguish newly arriving plant disease categories from previous ones, thereby optimising the prototype space and enhancing the recognition accuracy of new categories. We evaluated our approach on the PlantVillage dataset, and the experimental results under both 5-way 5-shot and 3-way 5-shot settings demonstrate that our method achieves state-of-the-art accuracy. At the same time, we achieved promising performance on the publicly available CIFAR-100 dataset. Furthermore, the visualisation results validate that our strategy effectively supports fine-grained, sustainable disease recognition, highlighting the potential of our approach to advance FSCIL in the field of plant disease monitoring.</p>\",\"PeriodicalId\":20100,\"journal\":{\"name\":\"Plant Methods\",\"volume\":\"21 1\",\"pages\":\"105\"},\"PeriodicalIF\":4.4000,\"publicationDate\":\"2025-07-31\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC12312506/pdf/\",\"citationCount\":\"0\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"Plant Methods\",\"FirstCategoryId\":\"99\",\"ListUrlMain\":\"https://doi.org/10.1186/s13007-025-01423-3\",\"RegionNum\":2,\"RegionCategory\":\"生物学\",\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"Q1\",\"JCRName\":\"BIOCHEMICAL RESEARCH METHODS\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"Plant Methods","FirstCategoryId":"99","ListUrlMain":"https://doi.org/10.1186/s13007-025-01423-3","RegionNum":2,"RegionCategory":"生物学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q1","JCRName":"BIOCHEMICAL RESEARCH METHODS","Score":null,"Total":0}
引用次数: 0

摘要

深度学习展示了强大的泛化能力,推动了植物病害识别系统的实质性进展。然而,当前的方法主要是针对离线实现进行优化的。实时作物监测系统遇到的流图像包含新疾病类别在少数条件下,需要增量自适应模型。这种能力被称为少次类增量学习(FSCIL)。在这里,我们引入了vcpv虚拟对比约束与原型载体校准,实现了FSClL条件下植物病害的可持续分类。具体来说,我们的方法包括两个阶段:基类训练阶段和增量训练阶段。在基类训练阶段,利用虚拟对比类约束(VCC)模块增强基类的学习,为新的植物病害图像分配足够的嵌入空间。在增量训练阶段,引入原型校准嵌入(PCE)模块来区分新到的植物病害类别,从而优化原型空间,提高新类别的识别精度。我们在PlantVillage数据集上评估了我们的方法,在5-way 5-shot和3-way 5-shot设置下的实验结果表明,我们的方法达到了最先进的精度。同时,我们在公开可用的CIFAR-100数据集上取得了很好的性能。此外,可视化结果验证了我们的策略有效地支持细粒度、可持续的疾病识别,突出了我们的方法在植物疾病监测领域推进FSCIL的潜力。
本文章由计算机程序翻译,如有差异,请以英文原文为准。
VCPC: virtual contrastive constraint and prototype calibration for few-shot class-incremental plant disease classification.

Deep learning demonstrates strong generalisation capabilities, driving substantial progress in plant disease recognition systems. However, current methods are predominantly optimised for offline implementation. Real-time crop surveillance systems encounter streaming images containing novel disease classes in few-shot conditions, demanding incrementally adaptive models. This capability is called few-shot class-incremental learning (FSCIL). Here, we introduce VCPV-virtual contrastive constraints with prototype vector calibration-enabling sustainable plant disease classification under FSClL conditions. Specifically, our method consists of two phases: the base class training phase and the incremental training phase. During the base class training phase, the virtual contrastive class constraints (VCC) module is utilised to enhance learning from base classes and allocate sufficient embedding space for new plant disease images. In the incremental training phase, the prototype calibration embedding (PCE) module is introduced to distinguish newly arriving plant disease categories from previous ones, thereby optimising the prototype space and enhancing the recognition accuracy of new categories. We evaluated our approach on the PlantVillage dataset, and the experimental results under both 5-way 5-shot and 3-way 5-shot settings demonstrate that our method achieves state-of-the-art accuracy. At the same time, we achieved promising performance on the publicly available CIFAR-100 dataset. Furthermore, the visualisation results validate that our strategy effectively supports fine-grained, sustainable disease recognition, highlighting the potential of our approach to advance FSCIL in the field of plant disease monitoring.

求助全文
通过发布文献求助,成功后即可免费获取论文全文。 去求助
来源期刊
Plant Methods
Plant Methods 生物-植物科学
CiteScore
9.20
自引率
3.90%
发文量
121
审稿时长
2 months
期刊介绍: Plant Methods is an open access, peer-reviewed, online journal for the plant research community that encompasses all aspects of technological innovation in the plant sciences. There is no doubt that we have entered an exciting new era in plant biology. The completion of the Arabidopsis genome sequence, and the rapid progress being made in other plant genomics projects are providing unparalleled opportunities for progress in all areas of plant science. Nevertheless, enormous challenges lie ahead if we are to understand the function of every gene in the genome, and how the individual parts work together to make the whole organism. Achieving these goals will require an unprecedented collaborative effort, combining high-throughput, system-wide technologies with more focused approaches that integrate traditional disciplines such as cell biology, biochemistry and molecular genetics. Technological innovation is probably the most important catalyst for progress in any scientific discipline. Plant Methods’ goal is to stimulate the development and adoption of new and improved techniques and research tools and, where appropriate, to promote consistency of methodologies for better integration of data from different laboratories.
×
引用
GB/T 7714-2015
复制
MLA
复制
APA
复制
导出至
BibTeX EndNote RefMan NoteFirst NoteExpress
×
提示
您的信息不完整,为了账户安全,请先补充。
现在去补充
×
提示
您因"违规操作"
具体请查看互助需知
我知道了
×
提示
确定
请完成安全验证×
copy
已复制链接
快去分享给好友吧!
我知道了
右上角分享
点击右上角分享
0
联系我们:info@booksci.cn Book学术提供免费学术资源搜索服务,方便国内外学者检索中英文文献。致力于提供最便捷和优质的服务体验。 Copyright © 2023 布克学术 All rights reserved.
京ICP备2023020795号-1
ghs 京公网安备 11010802042870号
Book学术文献互助
Book学术文献互助群
群 号:604180095
Book学术官方微信