Article(id=1317882353167479691, tenantId=1146029695717560320, journalId=1281212937352253451, issueId=1317882300113735707, articleNumber=null, orderNo=null, doi=10.12133/j.smartag.SA202504003, pmid=null, cstr=null, oa=null, hot=null, price=null, onlineType=0, articleFormat=0, articleType=null, articleTypeStr=research-article, receivedDate=1743523200000, receivedDateStr=2025-04-02, revisedDate=null, revisedDateStr=null, acceptedDate=null, acceptedDateStr=null, onlineDate=1792137238064, onlineDateStr=2026-10-16, pubDate=1769702400000, pubDateStr=2026-01-30, doiRegisterDate=null, doiRegisterDateStr=null, onlineIssueDate=1792137238064, onlineIssueDateStr=2026-10-16, onlineJustAcceptDate=null, onlineJustAcceptDateStr=null, onlineFirstDate=null, onlineFirstDateStr=null, sourceXml=null, magXml=null, createTime=1792137238064, creator=13701087609, updateTime=1792137238064, updator=13701087609, issue=Issue{id=1317882300113735707, tenantId=1146029695717560320, journalId=1281212937352253451, year='2026', volume='8', issue='1', pageStart='1', pageEnd='236', issueExtLink='null', onlineDate='null', pubDate='1769702400000', pubDateStr='2026-01-30', beforeIssueId=null, nextIssueId=null, price=null, status=1, issueComplete=1, articleOrder=1, issueType=-1, specialIssue=null, createTime=1792137225416, creator='13701087609', updateTime=1792137292328, updator='13701087609', preIssue=null, nextIssue=null, articleTotal=null, ext={EN=IssueExt(id=1317882580817527622, tenantId=1146029695717560320, journalId=1281212937352253451, issueId=1317882300113735707, language=EN, specialIssueTitle=, coverIllustrator=null, specialIssueEditor=, specialIssueAbout=), CN=IssueExt(id=1317882580817527623, tenantId=1146029695717560320, journalId=1281212937352253451, issueId=1317882300113735707, language=CN, specialIssueTitle=, coverIllustrator=null, specialIssueEditor=, specialIssueAbout=)}, issueFiles=null, downloadFileDto=null}, startPage=40, endPage=51, ext={EN=ArticleExt(id=1317882353364611980, articleId=1317882353167479691, tenantId=1146029695717560320, journalId=1281212937352253451, language=EN, title=Low-rank Adaptation Method for Fine-tuning Plant Disease Recognition Models, columnId=1317882300847738908, journalTitle=Smart Agriculture, columnName=Topic--Intelligent Identification and Diagnosis of Agricultural Diseases and Pests, runingTitle=null, highlight=null, articleAbstract=
[Objective] When deep learning is applied to plant disease recognition tasks, model fine-tuning faces significant challenges, including limited computational resources and high parameter update overhead. Although traditional low-rank adaptation (LoRA) methods effectively reduce parameter overhead, their strategy of assigning a uniform, fixed rank to all layers often overlooks the varying importance of different layers. This approach may still lead to constrained optimization in critical layers or resource waste in less significant ones. To address this limitation, a dynamic rank allocation (DRA) algorithm is proposed in this research. The DRA algorithm is designed to evaluate and adjust the required parameter resources for each layer during training, enhance the accuracy of plant disease classification models while more efficiently balancing computational resources. [Methods] Public datasets of the Wheat Plant Diseases Dataset and the Plants Disease Dataset were utilized in the experiments. The Wheat Plant Diseases Dataset comprised 13 104 images covering 15 types of wheat diseases such as black rust and fusarium head blight, while the Plants Disease Dataset included 37 505 images of 26 types of plant diseases such as algal leaf spot, corn rust, and bacterial spot of tomato. These datasets were captured under varied lighting, different backgrounds, diverse angles, and at various stages of plant growth. A cross-layer feature similarity metric based on centred kernel alignment (CKA) was introduced to quantify the representational correlation between different layers. Concurrently, a correction factor was constructed based on gradient information and activation intensity to measure the direct impact of each layer on the loss function. These two metrics were then fused using a weighted harmonic mean to generate a comprehensive importance score, which was subsequently used for the initial rank allocation. Furthermore, considering the effect of feature representation changes during training, a stability-triggered adaptive rank update strategy rank re-allocation (RRA) was proposed. This strategy monitored the average parameter change of the low-rank adapters during the training process to determine the convergence state. When this change fell below a specific threshold, the low-rank matrices were merged into the original weights, and the rank allocation table was then re-calculated and updated. This process ensured that more resources were allocated to critical layers, thereby achieving an optimized allocation of parameter resources across different layers. [Results and Discussions] Tests on four models (AlexNet, MobileNetV2, RegNetY, and ConvNeXt) indicated that, compared to full-parameter fine-tuning, the proposed method reduced resource consumption to 0.42%, 2.46%, 3.56%, and 1.25%, respectively, while maintaining a comparable average accuracy. The RRA strategy demonstrated continuous parameter optimization throughout the model's training. On the ConvNeXt model, the trainable parameters on the plants disease dataset were progressively reduced from 18.34 to 9.26 M, a reduction of nearly 50%. In comparison with the standard LoRA method (R=16), the method reduced the accuracy by 0.38, 0.40 and 0.05 percentage points on the wheat plant diseases dataset for AlexNet, MobileNetV2, and RegNetY, respectively, while resource consumption was reduced by 59.3%, 87.4% and 50.5%. Robustness was tested by applying perturbations to the test set, including Gaussian noise, random cropping, color jitter, and random rotation. The results showed that the model was most affected by color jitter and random rotation on the Plants Disease Dataset, with accuracy decreasing by 6.02 and 5.11 percentage points, respectively. On the wheat plant diseases dataset, the model was more sensitive to random cropping and random rotation, with accuracy decreasing by 4.33 and 4.40 percentage points, respectively; the overall performance degradation remained within an acceptable range. When compared to other advanced low-rank methods such as AdaLoRA and DyLoRA under the same parameter budget, the DRA method exhibited higher accuracy. On the RegNetY model, the DRA method achieved an accuracy of 90.96% on the Plants Disease Dataset, which was 0.55 percentage points higher than AdaLoRA and 0.94 percentage points higher than DyLoRA. In terms of training efficiency on the Plants Disease Dataset, the DRA method required 43.5 minutes to reach its peak validation accuracy of 89.84%, whereas AdaLoRA required 52.3 minutes, representing a training time increase of approximately 20.23%. Regarding inference flexibility, the DyLoRA method was designed to generate a universal model capable of adapting to multiple rank configurations after a single training run, allowing for dynamic rank switching during inference based on hardware or latency requirements. The DRA method, however, did not possess this inference-time flexibility. It was focused on converging to a single, high-performance rank configuration for a specific task during the training phase. [Conclusions] The low-rank adaptive fine-tuning method proposed in this research significantly reduced the number of model training parameters while ensuring plant disease recognition accuracy. Compared to traditional fixed-rank LoRA and other advanced low-rank optimization methods, it demonstrated distinct advantages, providing an effective pathway for efficient model deployment on resource-constrained devices.
, authors=Jinqing HUANG
1, 2, Jin YE
1, 2, Huilin HU
1, 2, Jihui YANG
3, Wei LAN
1, 2, Yanqing ZHANG
1, 2, authorsList=Jinqing HUANG, Jin YE, Huilin HU, Jihui YANG, Wei LAN, Yanqing ZHANG, authorCompany=null, correspAuthors=Jin YE, authorNote=
, correspAuthorsNote=
, copyrightStatement=copyright©2026 by the authors, copyrightOwner=null, extLink=null, articleAbsUrl=null, sourceXml=null, magXml=null, pdfUrl=null, pdf=null, pdfFileSize=null, pdfExtLink=null, richHtmlUrl=null, mobilePdfUrl=null, reviewReport=null, pdfFirstPage=null, abstractGraph=null, abstractGraphContent=null, abstractVideo=null, citation=null, cebUrl=null, magXmlContent=null, mapNumber=null, fund=null), CN=ArticleExt(id=1317882354803258263, articleId=1317882353167479691, tenantId=1146029695717560320, journalId=1281212937352253451, language=CN, title=植物病害识别微调模型低秩适配方法, columnId=1317882301023899678, journalTitle=智慧农业(中英文), columnName=专题--农业病虫害智能识别与诊断, runingTitle=null, highlight=null, articleAbstract=
【目的/意义】 深度学习应用于植物病害识别任务时,模型微调面临计算资源受限、参数更新开销较大的挑战。传统低秩适配(Low-Rank Adaptation, LoRA)方法虽能有效减少参数开销,但仍可能导致资源浪费或优化受限。 【方法】 提出一种动态秩分配算法,旨在提升植物病害分类模型精度的同时,平衡计算资源。通过引入一种基于中心化核对齐的跨层特征相似性度量方法,量化不同层之间的表征相关性;同时基于梯度信息和激活强度构建1个校正因子,用于衡量各层对损失函数的直接影响;考虑训练过程中特征表示变化的影响,提出基于稳定性触发的自适应秩分配表更新策略(Rank Re-Allocation, RRA),在训练过程中监测微调模型参数的收敛状态,在模型趋于稳定时自动更新秩分配表,确保关键层分配更多资源,从而实现参数资源在不同层之间的优化分配。 【结果和讨论】 在小麦病害数据集、植物病害数据集等公开数据集上使用AlexNet、MobileNetV2、RegNetY和ConvNeXt这4种不同的模型进行测试,该方法与全参数微调相比,在保证平均准确率的基础上,资源消耗分别减少至0.42%,2.46%,3.56%和1.25%。与LoRA方法(R=16)进行对比,AlexNet、MobileNetV2和RegNetY在小麦病害数据集上的准确率分别提高0.38、0.40和0.05个百分点,资源消耗分别减少59.3%、87.4%和50.5%。 【结论】 低秩自适应微调方法在保证植物病害识别精度的同时,能够显著减少模型训练参数。
, authors=黄锦清
1, 2, 叶进
1, 2, 胡慧琳
1, 2, 阳继辉
3, 兰伟
1, 2, 张艳青
1, 2, authorsList=黄锦清, 叶进, 胡慧琳, 阳继辉, 兰伟, 张艳青, authorCompany=null, correspAuthors=叶进, authorNote=
, correspAuthorsNote=
, copyrightStatement=null, copyrightOwner=null, extLink=null, articleAbsUrl=null, sourceXml=NscP4GnX+GCnD80QVVQ6OQ==, magXml=0LuAJnzuwl0BA53tccLGaw==, pdfUrl=null, pdf=ErnhA8YTwHNX8VwJrriiqg==, pdfFileSize=2303006, pdfExtLink=null, richHtmlUrl=null, mobilePdfUrl=null, reviewReport=null, pdfFirstPage=null, abstractGraph=lQfQfxEDBmhb2s/P7tDkGw==, abstractGraphContent=null, abstractVideo=null, citation=null, cebUrl=null, magXmlContent=N6rC5x4GZoP3pS0gScG9pg==, mapNumber=null, fund=null)}, authors=[Author(id=1317882750804279963, tenantId=1146029695717560320, journalId=1281212937352253451, articleId=1317882353167479691, orderNo=0, firstName=null, middleName=null, lastName=null, nameCn=null, orcid=null, stid=null, country=null, authorPic=null, dead=0, email=1766042592@qq.com, emailSecond=null, emailThird=null, correspondingAuthor=0, authorType=1, ext={EN=AuthorExt(id=1317882750883971742, tenantId=1146029695717560320, journalId=1281212937352253451, articleId=1317882353167479691, authorId=1317882750804279963, language=EN, stringName=Jinqing HUANG, firstName=Jinqing, middleName=null, lastName=HUANG, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=
1, 2, address=
1.School of Computer, Electronics and Information, Guangxi University, Nanning 530000, China
2.State Key Laboratory for Conservation and Utilization of Subtropical Agrobiological Resources, Nanning 530000, China, bio={"content":"
HUANG Jinqing, E-mail: 1766042592@qq.com
"}, bioImg=null, bioContent=
HUANG Jinqing, E-mail: 1766042592@qq.com
, aboutCorrespAuthor=null), CN=AuthorExt(id=1317882750951080607, tenantId=1146029695717560320, journalId=1281212937352253451, articleId=1317882353167479691, authorId=1317882750804279963, language=CN, stringName=黄锦清, firstName=null, middleName=null, lastName=null, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=
1, 2, address=
1.广西大学 计算机与电子信息学院,广西 南宁 530000,中国
2.亚热带农业保护与利用国家重点实验室,广西 南宁 530000,中国, bio={"content":"
黄锦清,研究方向为大模型微调。E-mail:1766042592@qq.com
"}, bioImg=null, bioContent=
黄锦清,研究方向为大模型微调。E-mail:1766042592@qq.com
, aboutCorrespAuthor=null)}, companyList=[AuthorCompany(id=1317882750472929937, tenantId=1146029695717560320, journalId=1281212937352253451, articleId=1317882353167479691, xref=1., ext=[AuthorCompanyExt(id=1317882750481318546, tenantId=1146029695717560320, journalId=1281212937352253451, articleId=1317882353167479691, companyId=1317882750472929937, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=
1.School of Computer, Electronics and Information, Guangxi University, Nanning 530000, China), AuthorCompanyExt(id=1317882750577787539, tenantId=1146029695717560320, journalId=1281212937352253451, articleId=1317882353167479691, companyId=1317882750472929937, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=
1.广西大学 计算机与电子信息学院,广西 南宁 530000,中国)]), AuthorCompany(id=1317882750644896404, tenantId=1146029695717560320, journalId=1281212937352253451, articleId=1317882353167479691, xref=2., ext=[AuthorCompanyExt(id=1317882750657479317, tenantId=1146029695717560320, journalId=1281212937352253451, articleId=1317882353167479691, companyId=1317882750644896404, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=
2.State Key Laboratory for Conservation and Utilization of Subtropical Agrobiological Resources, Nanning 530000, China), AuthorCompanyExt(id=1317882750665867926, tenantId=1146029695717560320, journalId=1281212937352253451, articleId=1317882353167479691, companyId=1317882750644896404, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=
2.亚热带农业保护与利用国家重点实验室,广西 南宁 530000,中国)])]), Author(id=1317882751018189473, tenantId=1146029695717560320, journalId=1281212937352253451, articleId=1317882353167479691, orderNo=1, firstName=null, middleName=null, lastName=null, nameCn=null, orcid=null, stid=null, country=null, authorPic=null, dead=0, email=yejin@gxu.edu.cn, emailSecond=null, emailThird=null, correspondingAuthor=1, authorType=1, ext={EN=AuthorExt(id=1317882751089492644, tenantId=1146029695717560320, journalId=1281212937352253451, articleId=1317882353167479691, authorId=1317882751018189473, language=EN, stringName=Jin YE, firstName=Jin, middleName=null, lastName=YE, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=
1, 2, address=
1.School of Computer, Electronics and Information, Guangxi University, Nanning 530000, China
2.State Key Laboratory for Conservation and Utilization of Subtropical Agrobiological Resources, Nanning 530000, China, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null), CN=AuthorExt(id=1317882751152407205, tenantId=1146029695717560320, journalId=1281212937352253451, articleId=1317882353167479691, authorId=1317882751018189473, language=CN, stringName=叶进, firstName=null, middleName=null, lastName=null, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=
1, 2, address=
1.广西大学 计算机与电子信息学院,广西 南宁 530000,中国
2.亚热带农业保护与利用国家重点实验室,广西 南宁 530000,中国, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null)}, companyList=[AuthorCompany(id=1317882750472929937, tenantId=1146029695717560320, journalId=1281212937352253451, articleId=1317882353167479691, xref=1., ext=[AuthorCompanyExt(id=1317882750481318546, tenantId=1146029695717560320, journalId=1281212937352253451, articleId=1317882353167479691, companyId=1317882750472929937, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=
1.School of Computer, Electronics and Information, Guangxi University, Nanning 530000, China), AuthorCompanyExt(id=1317882750577787539, tenantId=1146029695717560320, journalId=1281212937352253451, articleId=1317882353167479691, companyId=1317882750472929937, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=
1.广西大学 计算机与电子信息学院,广西 南宁 530000,中国)]), AuthorCompany(id=1317882750644896404, tenantId=1146029695717560320, journalId=1281212937352253451, articleId=1317882353167479691, xref=2., ext=[AuthorCompanyExt(id=1317882750657479317, tenantId=1146029695717560320, journalId=1281212937352253451, articleId=1317882353167479691, companyId=1317882750644896404, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=
2.State Key Laboratory for Conservation and Utilization of Subtropical Agrobiological Resources, Nanning 530000, China), AuthorCompanyExt(id=1317882750665867926, tenantId=1146029695717560320, journalId=1281212937352253451, articleId=1317882353167479691, companyId=1317882750644896404, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=
2.亚热带农业保护与利用国家重点实验室,广西 南宁 530000,中国)])]), Author(id=1317882751215321767, tenantId=1146029695717560320, journalId=1281212937352253451, articleId=1317882353167479691, orderNo=2, firstName=null, middleName=null, lastName=null, nameCn=null, orcid=null, stid=null, country=null, authorPic=null, dead=0, email=null, emailSecond=null, emailThird=null, correspondingAuthor=0, authorType=1, ext={EN=AuthorExt(id=1317882751290819242, tenantId=1146029695717560320, journalId=1281212937352253451, articleId=1317882353167479691, authorId=1317882751215321767, language=EN, stringName=Huilin HU, firstName=Huilin, middleName=null, lastName=HU, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=
1, 2, address=
1.School of Computer, Electronics and Information, Guangxi University, Nanning 530000, China
2.State Key Laboratory for Conservation and Utilization of Subtropical Agrobiological Resources, Nanning 530000, China, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null), CN=AuthorExt(id=1317882751370511019, tenantId=1146029695717560320, journalId=1281212937352253451, articleId=1317882353167479691, authorId=1317882751215321767, language=CN, stringName=胡慧琳, firstName=null, middleName=null, lastName=null, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=
1, 2, address=
1.广西大学 计算机与电子信息学院,广西 南宁 530000,中国
2.亚热带农业保护与利用国家重点实验室,广西 南宁 530000,中国, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null)}, companyList=[AuthorCompany(id=1317882750472929937, tenantId=1146029695717560320, journalId=1281212937352253451, articleId=1317882353167479691, xref=1., ext=[AuthorCompanyExt(id=1317882750481318546, tenantId=1146029695717560320, journalId=1281212937352253451, articleId=1317882353167479691, companyId=1317882750472929937, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=
1.School of Computer, Electronics and Information, Guangxi University, Nanning 530000, China), AuthorCompanyExt(id=1317882750577787539, tenantId=1146029695717560320, journalId=1281212937352253451, articleId=1317882353167479691, companyId=1317882750472929937, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=
1.广西大学 计算机与电子信息学院,广西 南宁 530000,中国)]), AuthorCompany(id=1317882750644896404, tenantId=1146029695717560320, journalId=1281212937352253451, articleId=1317882353167479691, xref=2., ext=[AuthorCompanyExt(id=1317882750657479317, tenantId=1146029695717560320, journalId=1281212937352253451, articleId=1317882353167479691, companyId=1317882750644896404, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=
2.State Key Laboratory for Conservation and Utilization of Subtropical Agrobiological Resources, Nanning 530000, China), AuthorCompanyExt(id=1317882750665867926, tenantId=1146029695717560320, journalId=1281212937352253451, articleId=1317882353167479691, companyId=1317882750644896404, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=
2.亚热带农业保护与利用国家重点实验室,广西 南宁 530000,中国)])]), Author(id=1317882751433425581, tenantId=1146029695717560320, journalId=1281212937352253451, articleId=1317882353167479691, orderNo=3, firstName=null, middleName=null, lastName=null, nameCn=null, orcid=null, stid=null, country=null, authorPic=null, dead=0, email=null, emailSecond=null, emailThird=null, correspondingAuthor=0, authorType=1, ext={EN=AuthorExt(id=1317882751534088879, tenantId=1146029695717560320, journalId=1281212937352253451, articleId=1317882353167479691, authorId=1317882751433425581, language=EN, stringName=Jihui YANG, firstName=Jihui, middleName=null, lastName=YANG, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=
3, address=
3.JJR Science and Technology Group Co. , Ltd. , Nanning 530000, China, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null), CN=AuthorExt(id=1317882751597003440, tenantId=1146029695717560320, journalId=1281212937352253451, articleId=1317882353167479691, authorId=1317882751433425581, language=CN, stringName=阳继辉, firstName=null, middleName=null, lastName=null, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=
3, address=
3.捷佳润科技集团股份有限公司,广西 南宁 530000,中国, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null)}, companyList=[AuthorCompany(id=1317882750728782487, tenantId=1146029695717560320, journalId=1281212937352253451, articleId=1317882353167479691, xref=3., ext=[AuthorCompanyExt(id=1317882750741365400, tenantId=1146029695717560320, journalId=1281212937352253451, articleId=1317882353167479691, companyId=1317882750728782487, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=
3.JJR Science and Technology Group Co. , Ltd. , Nanning 530000, China), AuthorCompanyExt(id=1317882750745559705, tenantId=1146029695717560320, journalId=1281212937352253451, articleId=1317882353167479691, companyId=1317882750728782487, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=
3.捷佳润科技集团股份有限公司,广西 南宁 530000,中国)])]), Author(id=1317882751659918002, tenantId=1146029695717560320, journalId=1281212937352253451, articleId=1317882353167479691, orderNo=4, firstName=null, middleName=null, lastName=null, nameCn=null, orcid=null, stid=null, country=null, authorPic=null, dead=0, email=null, emailSecond=null, emailThird=null, correspondingAuthor=0, authorType=1, ext={EN=AuthorExt(id=1317882751743804085, tenantId=1146029695717560320, journalId=1281212937352253451, articleId=1317882353167479691, authorId=1317882751659918002, language=EN, stringName=Wei LAN, firstName=Wei, middleName=null, lastName=LAN, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=
1, 2, address=
1.School of Computer, Electronics and Information, Guangxi University, Nanning 530000, China
2.State Key Laboratory for Conservation and Utilization of Subtropical Agrobiological Resources, Nanning 530000, China, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null), CN=AuthorExt(id=1317882753526383287, tenantId=1146029695717560320, journalId=1281212937352253451, articleId=1317882353167479691, authorId=1317882751659918002, language=CN, stringName=兰伟, firstName=null, middleName=null, lastName=null, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=
1, 2, address=
1.广西大学 计算机与电子信息学院,广西 南宁 530000,中国
2.亚热带农业保护与利用国家重点实验室,广西 南宁 530000,中国, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null)}, companyList=[AuthorCompany(id=1317882750472929937, tenantId=1146029695717560320, journalId=1281212937352253451, articleId=1317882353167479691, xref=1., ext=[AuthorCompanyExt(id=1317882750481318546, tenantId=1146029695717560320, journalId=1281212937352253451, articleId=1317882353167479691, companyId=1317882750472929937, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=
1.School of Computer, Electronics and Information, Guangxi University, Nanning 530000, China), AuthorCompanyExt(id=1317882750577787539, tenantId=1146029695717560320, journalId=1281212937352253451, articleId=1317882353167479691, companyId=1317882750472929937, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=
1.广西大学 计算机与电子信息学院,广西 南宁 530000,中国)]), AuthorCompany(id=1317882750644896404, tenantId=1146029695717560320, journalId=1281212937352253451, articleId=1317882353167479691, xref=2., ext=[AuthorCompanyExt(id=1317882750657479317, tenantId=1146029695717560320, journalId=1281212937352253451, articleId=1317882353167479691, companyId=1317882750644896404, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=
2.State Key Laboratory for Conservation and Utilization of Subtropical Agrobiological Resources, Nanning 530000, China), AuthorCompanyExt(id=1317882750665867926, tenantId=1146029695717560320, journalId=1281212937352253451, articleId=1317882353167479691, companyId=1317882750644896404, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=
2.亚热带农业保护与利用国家重点实验室,广西 南宁 530000,中国)])]), Author(id=1317882753614463673, tenantId=1146029695717560320, journalId=1281212937352253451, articleId=1317882353167479691, orderNo=5, firstName=null, middleName=null, lastName=null, nameCn=null, orcid=null, stid=null, country=null, authorPic=null, dead=0, email=null, emailSecond=null, emailThird=null, correspondingAuthor=0, authorType=1, ext={EN=AuthorExt(id=1317882753698349756, tenantId=1146029695717560320, journalId=1281212937352253451, articleId=1317882353167479691, authorId=1317882753614463673, language=EN, stringName=Yanqing ZHANG, firstName=Yanqing, middleName=null, lastName=ZHANG, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=
1, 2, address=
1.School of Computer, Electronics and Information, Guangxi University, Nanning 530000, China
2.State Key Laboratory for Conservation and Utilization of Subtropical Agrobiological Resources, Nanning 530000, China, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null), CN=AuthorExt(id=1317882753941619389, tenantId=1146029695717560320, journalId=1281212937352253451, articleId=1317882353167479691, authorId=1317882753614463673, language=CN, stringName=张艳青, firstName=null, middleName=null, lastName=null, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=
1, 2, address=
1.广西大学 计算机与电子信息学院,广西 南宁 530000,中国
2.亚热带农业保护与利用国家重点实验室,广西 南宁 530000,中国, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null)}, companyList=[AuthorCompany(id=1317882750472929937, tenantId=1146029695717560320, journalId=1281212937352253451, articleId=1317882353167479691, xref=1., ext=[AuthorCompanyExt(id=1317882750481318546, tenantId=1146029695717560320, journalId=1281212937352253451, articleId=1317882353167479691, companyId=1317882750472929937, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=
1.School of Computer, Electronics and Information, Guangxi University, Nanning 530000, China), AuthorCompanyExt(id=1317882750577787539, tenantId=1146029695717560320, journalId=1281212937352253451, articleId=1317882353167479691, companyId=1317882750472929937, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=
1.广西大学 计算机与电子信息学院,广西 南宁 530000,中国)]), AuthorCompany(id=1317882750644896404, tenantId=1146029695717560320, journalId=1281212937352253451, articleId=1317882353167479691, xref=2., ext=[AuthorCompanyExt(id=1317882750657479317, tenantId=1146029695717560320, journalId=1281212937352253451, articleId=1317882353167479691, companyId=1317882750644896404, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=
2.State Key Laboratory for Conservation and Utilization of Subtropical Agrobiological Resources, Nanning 530000, China), AuthorCompanyExt(id=1317882750665867926, tenantId=1146029695717560320, journalId=1281212937352253451, articleId=1317882353167479691, companyId=1317882750644896404, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=
2.亚热带农业保护与利用国家重点实验室,广西 南宁 530000,中国)])])], keywords=[Keyword(id=1317882754482684606, tenantId=1146029695717560320, journalId=1281212937352253451, articleId=1317882353167479691, language=EN, orderNo=1, keyword=low-rank adaptive fine-tuning), Keyword(id=1317882754558182079, tenantId=1146029695717560320, journalId=1281212937352253451, articleId=1317882353167479691, language=EN, orderNo=2, keyword=feature similarity), Keyword(id=1317882754839200448, tenantId=1146029695717560320, journalId=1281212937352253451, articleId=1317882353167479691, language=EN, orderNo=3, keyword=dynamic rank allocation algorithm), Keyword(id=1317882754910503617, tenantId=1146029695717560320, journalId=1281212937352253451, articleId=1317882353167479691, language=EN, orderNo=4, keyword=rank allocation update strategy), Keyword(id=1317882755191521986, tenantId=1146029695717560320, journalId=1281212937352253451, articleId=1317882353167479691, language=CN, orderNo=1, keyword=低秩自适应微调), Keyword(id=1317882755254436547, tenantId=1146029695717560320, journalId=1281212937352253451, articleId=1317882353167479691, language=CN, orderNo=2, keyword=特征相似性), Keyword(id=1317882755334128324, tenantId=1146029695717560320, journalId=1281212937352253451, articleId=1317882353167479691, language=CN, orderNo=3, keyword=秩分配算法), Keyword(id=1317882755610952389, tenantId=1146029695717560320, journalId=1281212937352253451, articleId=1317882353167479691, language=CN, orderNo=4, keyword=秩分配更新策略)], refs=[Reference(id=1317882762959373031, tenantId=1146029695717560320, journalId=1281212937352253451, articleId=1317882353167479691, doi=null, pmid=null, pmcid=null, year=2019, volume=1, issue=3, pageStart=46, pageEnd=55, url=null, language=null, rfNumber=[1], rfOrder=0, authorNames=李淼, 王敬贤, 李华龙, journalName=智慧农业, refType=null, unstructuredReference=李淼, 王敬贤, 李华龙,
等. 基于CNN和迁移学习的农作物病害识别方法研究[J].
智慧农业,
2019,
1(3): 46-55., articleTitle=基于CNN和迁移学习的农作物病害识别方法研究, refAbstract=null), Reference(id=1317882763173282536, tenantId=1146029695717560320, journalId=1281212937352253451, articleId=1317882353167479691, doi=null, pmid=null, pmcid=null, year=2019, volume=1, issue=3, pageStart=46, pageEnd=55, url=null, language=null, rfNumber=[1], rfOrder=1, authorNames=LI M, WANG J X, LI H L, journalName=Smart Agriculture, refType=null, unstructuredReference=
LI M,
WANG J X,
LI H L,
et al. Method for identifying crop disease based on CNN and transfer learning[J].
Smart Agriculture,
2019,
1(3): 46-55., articleTitle=null, refAbstract=null), Reference(id=1317882763261362921, tenantId=1146029695717560320, journalId=1281212937352253451, articleId=1317882353167479691, doi=null, pmid=null, pmcid=null, year=2025, volume=56, issue=4, pageStart=93, pageEnd=102, url=null, language=null, rfNumber=[2], rfOrder=2, authorNames=乔世成, 潘春宇, 白明宇, journalName=沈阳农业大学学报, refType=null, unstructuredReference=乔世成, 潘春宇, 白明宇,
等. 基于改进YOLOv10n的石榴病害检测[J].
沈阳农业大学学报,
2025,
56(4): 93-102., articleTitle=基于改进YOLOv10n的石榴病害检测, refAbstract=null), Reference(id=1317882763336860396, tenantId=1146029695717560320, journalId=1281212937352253451, articleId=1317882353167479691, doi=null, pmid=null, pmcid=null, year=2025, volume=56, issue=4, pageStart=93, pageEnd=102, url=null, language=null, rfNumber=[2], rfOrder=3, authorNames=QIAO S C, PAN C Y, BAI M Y, journalName=Journal of Shenyang Agricultural University, refType=null, unstructuredReference=
QIAO S C,
PAN C Y,
BAI M Y,
et al. Pomegranate disease detection based on improved YOLOv10n[J].
Journal of Shenyang Agricultural University,
2025,
56(4): 93-102., articleTitle=null, refAbstract=null), Reference(id=1317882763412357869, tenantId=1146029695717560320, journalId=1281212937352253451, articleId=1317882353167479691, doi=null, pmid=null, pmcid=null, year=2017, volume=60, issue=6, pageStart=84, pageEnd=90, url=null, language=null, rfNumber=[3], rfOrder=4, authorNames=KRIZHEVSKY A, SUTSKEVER I, HINTON G E, journalName=Communications of The Acm, refType=null, unstructuredReference=
KRIZHEVSKY A,
SUTSKEVER I,
HINTON G E. ImageNet classification with deep convolutional neural networks[J].
Communications of The Acm,
2017,
60(6): 84-90., articleTitle=ImageNet classification with deep convolutional neural networks, refAbstract=null), Reference(id=1317882763647238894, tenantId=1146029695717560320, journalId=1281212937352253451, articleId=1317882353167479691, doi=null, pmid=null, pmcid=null, year=2024, volume=223, issue=null, pageStart=109090, pageEnd=null, url=null, language=null, rfNumber=[4], rfOrder=5, authorNames=BADGUJAR C M, POULOSE A, GAN H, journalName=Computers and Electronics in Agriculture, refType=null, unstructuredReference=
BADGUJAR C M,
POULOSE A,
GAN H. Agricultural object detection with You Only Look Once (YOLO) Algorithm: A bibliometric and systematic literature review[J].
Computers and Electronics in Agriculture,
2024,
223: 109090., articleTitle=Agricultural object detection with You Only Look Once (YOLO) Algorithm: A bibliometric and systematic literature review, refAbstract=null), Reference(id=1317882763722736367, tenantId=1146029695717560320, journalId=1281212937352253451, articleId=1317882353167479691, doi=null, pmid=null, pmcid=null, year=2016, volume=null, issue=null, pageStart=770, pageEnd=778, url=null, language=null, rfNumber=[5], rfOrder=6, authorNames=HE K M, ZHANG X Y, REN S Q, journalName=null, refType=null, unstructuredReference=
HE K M,
ZHANG X Y,
REN S Q,
et al. Deep residual learning for image recognition[C]// 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR). Piscataway, New Jersey, USA: IEEE,
2016: 770-778., articleTitle=Deep residual learning for image recognition, refAbstract=null), Reference(id=1317882764058280688, tenantId=1146029695717560320, journalId=1281212937352253451, articleId=1317882353167479691, doi=null, pmid=null, pmcid=null, year=2023, volume=5, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=[6], rfOrder=7, authorNames=DONG X Y, WANG Q, HUANG Q D, journalName=Plant Phenomics, refType=null, unstructuredReference=
DONG X Y,
WANG Q,
HUANG Q D,
et al. PDDD-PreTrain: A series of commonly used pre-trained models support image-based plant disease diagnosis[J].
Plant Phenomics,
2023,
5: 0054., articleTitle=PDDD-PreTrain: A series of commonly used pre-trained models support image-based plant disease diagnosis, refAbstract=null), Reference(id=1317882764142166769, tenantId=1146029695717560320, journalId=1281212937352253451, articleId=1317882353167479691, doi=null, pmid=null, pmcid=null, year=2022, volume=null, issue=null, pageStart=11966, pageEnd=11976, url=null, language=null, rfNumber=[7], rfOrder=8, authorNames=LIU Z, MAO H Z, WU C Y, journalName=null, refType=null, unstructuredReference=
LIU Z,
MAO H Z,
WU C Y,
et al. A ConvNet for the 2020s[C]// 2022 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR). Piscataway, New Jersey, USA: IEEE,
2022: 11966-11976., articleTitle=A ConvNet for the 2020s, refAbstract=null), Reference(id=1317882764209275634, tenantId=1146029695717560320, journalId=1281212937352253451, articleId=1317882353167479691, doi=null, pmid=null, pmcid=null, year=2019, volume=null, issue=null, pageStart=1314, pageEnd=1324, url=null, language=null, rfNumber=[8], rfOrder=9, authorNames=HOWARD A, SANDLER M, CHEN B, journalName=null, refType=null, unstructuredReference=
HOWARD A,
SANDLER M,
CHEN B,
et al. Searching for MobileNetV3[C]// 2019 IEEE/CVF International Conference on Computer Vision (ICCV). Piscataway, New Jersey, USA: IEEE,
2019: 1314-1324., articleTitle=Searching for MobileNetV3, refAbstract=null), Reference(id=1317882764280578803, tenantId=1146029695717560320, journalId=1281212937352253451, articleId=1317882353167479691, doi=null, pmid=null, pmcid=null, year=2018, volume=145, issue=null, pageStart=311, pageEnd=318, url=null, language=null, rfNumber=[9], rfOrder=10, authorNames=FERENTINOS K P, journalName=Computers and Electronics in Agriculture, refType=null, unstructuredReference=
FERENTINOS K P. Deep learning models for plant disease detection and diagnosis[J].
Computers and Electronics in Agriculture,
2018,
145: 311-318., articleTitle=Deep learning models for plant disease detection and diagnosis, refAbstract=null), Reference(id=1317882764477711092, tenantId=1146029695717560320, journalId=1281212937352253451, articleId=1317882353167479691, doi=null, pmid=null, pmcid=null, year=2024, volume=24, issue=12, pageStart=4811, pageEnd=4823, url=null, language=null, rfNumber=[10], rfOrder=11, authorNames=于明, 郭志永, 王岩, journalName=科学技术与工程, refType=null, unstructuredReference=于明, 郭志永, 王岩. 基于计算机视觉的植物病害识别方法综述[J].
科学技术与工程,
2024,
24(12): 4811-4823., articleTitle=基于计算机视觉的植物病害识别方法综述, refAbstract=null), Reference(id=1317882764553208565, tenantId=1146029695717560320, journalId=1281212937352253451, articleId=1317882353167479691, doi=null, pmid=null, pmcid=null, year=2024, volume=24, issue=12, pageStart=4811, pageEnd=4823, url=null, language=null, rfNumber=[10], rfOrder=12, authorNames=YU M, GUO Z Y, WANG Y, journalName=Science Technology and Engineering, refType=null, unstructuredReference=
YU M,
GUO Z Y,
WANG Y. Review of computer vision-based plant disease identification techniques[J].
Science Technology and Engineering,
2024,
24(12): 4811-4823., articleTitle=null, refAbstract=null), Reference(id=1317882764624511734, tenantId=1146029695717560320, journalId=1281212937352253451, articleId=1317882353167479691, doi=null, pmid=null, pmcid=null, year=2024, volume=6, issue=6, pageStart=144, pageEnd=154, url=null, language=null, rfNumber=[11], rfOrder=13, authorNames=傅卓军, 胡政, 邓阳君, journalName=智慧农业(中英文), refType=null, unstructuredReference=傅卓军, 胡政, 邓阳君,
等. 基于Deep-Semi-NMF的苹果斑点落叶病检测方法[J].
智慧农业(中英文),
2024,
6(6): 144-154., articleTitle=基于Deep-Semi-NMF的苹果斑点落叶病检测方法, refAbstract=null), Reference(id=1317882764884558583, tenantId=1146029695717560320, journalId=1281212937352253451, articleId=1317882353167479691, doi=null, pmid=null, pmcid=null, year=2024, volume=6, issue=6, pageStart=144, pageEnd=154, url=null, language=null, rfNumber=[11], rfOrder=14, authorNames=FU Z J, HU Z, DENG Y J, journalName=Smart Agriculture, refType=null, unstructuredReference=
FU Z J,
HU Z,
DENG Y J,
et al. Detection method of apple Alternaria leaf spot based on deep-semi-NMF[J].
Smart Agriculture,
2024,
6(6): 144-154., articleTitle=null, refAbstract=null), Reference(id=1317882764964250360, tenantId=1146029695717560320, journalId=1281212937352253451, articleId=1317882353167479691, doi=null, pmid=null, pmcid=null, year=2025, volume=7, issue=3, pageStart=294, pageEnd=306, url=null, language=null, rfNumber=[12], rfOrder=15, authorNames=康继昌, 赵连军, journalName=农业大数据学报, refType=null, unstructuredReference=康继昌, 赵连军. 基于PSA-YOLO11n的小麦害虫检测[J].
农业大数据学报,
2025,
7(3): 294-306., articleTitle=基于PSA-YOLO11n的小麦害虫检测, refAbstract=null), Reference(id=1317882765043942137, tenantId=1146029695717560320, journalId=1281212937352253451, articleId=1317882353167479691, doi=null, pmid=null, pmcid=null, year=2025, volume=7, issue=3, pageStart=294, pageEnd=306, url=null, language=null, rfNumber=[12], rfOrder=16, authorNames=KANG J C, ZHAO L J, journalName=Journal of Agricultural Big Data, refType=null, unstructuredReference=
KANG J C,
ZHAO L J. Wheat pest detection based on PSA-YOLO11n[J].
Journal of Agricultural Big Data,
2025,
7(3): 294-306., articleTitle=null, refAbstract=null), Reference(id=1317882765262045946, tenantId=1146029695717560320, journalId=1281212937352253451, articleId=1317882353167479691, doi=null, pmid=null, pmcid=null, year=2023, volume=39, issue=14, pageStart=184, pageEnd=190, url=null, language=null, rfNumber=[13], rfOrder=17, authorNames=李想, 胡肖楠, 李方一, journalName=农业工程学报, refType=null, unstructuredReference=李想, 胡肖楠, 李方一,
等. 苹果树叶多病害及不可辨别病害的轻量识别算法[J].
农业工程学报,
2023,
39(14): 184-190., articleTitle=苹果树叶多病害及不可辨别病害的轻量识别算法, refAbstract=null), Reference(id=1317882766948156155, tenantId=1146029695717560320, journalId=1281212937352253451, articleId=1317882353167479691, doi=null, pmid=null, pmcid=null, year=2023, volume=39, issue=14, pageStart=184, pageEnd=190, url=null, language=null, rfNumber=[13], rfOrder=18, authorNames=LI X, HU X N, LI F Y, journalName=Transactions of the Chinese Society of Agricultural Engineering, refType=null, unstructuredReference=
LI X,
HU X N,
LI F Y,
et al. Lightweight recognition for multiple and indistinguishable diseases of apple tree leaf[J].
Transactions of the Chinese Society of Agricultural Engineering,
2023,
39(14): 184-190., articleTitle=null, refAbstract=null), Reference(id=1317882767023653628, tenantId=1146029695717560320, journalId=1281212937352253451, articleId=1317882353167479691, doi=null, pmid=null, pmcid=null, year=2023, volume=54, issue=5, pageStart=253, pageEnd=260, url=null, language=null, rfNumber=[14], rfOrder=19, authorNames=叶进, 吴梦岚, 邱文杰, journalName=农业机械学报, refType=null, unstructuredReference=叶进, 吴梦岚, 邱文杰,
等. 基于polyphyletic损失函数的荔枝花检测方法[J].
农业机械学报,
2023,
54(5): 253-260., articleTitle=基于polyphyletic损失函数的荔枝花检测方法, refAbstract=null), Reference(id=1317882767346615037, tenantId=1146029695717560320, journalId=1281212937352253451, articleId=1317882353167479691, doi=null, pmid=null, pmcid=null, year=2023, volume=54, issue=5, pageStart=253, pageEnd=260, url=null, language=null, rfNumber=[14], rfOrder=20, authorNames=YE J, WU M L, QIU W J, journalName=Transactions of the Chinese Society for Agricultural Machinery, refType=null, unstructuredReference=
YE J,
WU M L,
QIU W J,
et al. Litchi flower detection method based on polyphyletic loss function[J].
Transactions of the Chinese Society for Agricultural Machinery,
2023,
54(5): 253-260., articleTitle=null, refAbstract=null), Reference(id=1317882767430501118, tenantId=1146029695717560320, journalId=1281212937352253451, articleId=1317882353167479691, doi=null, pmid=null, pmcid=null, year=2021, volume=3, issue=1, pageStart=109, pageEnd=117, url=null, language=null, rfNumber=[15], rfOrder=21, authorNames=邱文杰, 叶进, 胡亮青, journalName=智慧农业(中英文), refType=null, unstructuredReference=邱文杰, 叶进, 胡亮青,
等. 面向植物病害识别的卷积神经网络精简结构Distilled-MobileNet模型[J].
智慧农业(中英文),
2021,
3(1): 109-117., articleTitle=面向植物病害识别的卷积神经网络精简结构Distilled-MobileNet模型, refAbstract=null), Reference(id=1317882767787016959, tenantId=1146029695717560320, journalId=1281212937352253451, articleId=1317882353167479691, doi=null, pmid=null, pmcid=null, year=2021, volume=3, issue=1, pageStart=109, pageEnd=117, url=null, language=null, rfNumber=[15], rfOrder=22, authorNames=QIU W J, YE J, HU L Q, journalName=Smart Agriculture, refType=null, unstructuredReference=
QIU W J,
YE J,
HU L Q,
et al. Distilled-MobileNet model of convolutional neural network simplified structure for plant disease recognition[J].
Smart Agriculture,
2021,
3(1): 109-117., articleTitle=null, refAbstract=null), Reference(id=1317882767858320128, tenantId=1146029695717560320, journalId=1281212937352253451, articleId=1317882353167479691, doi=null, pmid=null, pmcid=null, year=2024, volume=6, issue=3, pageStart=46, pageEnd=57, url=null, language=null, rfNumber=[16], rfOrder=23, authorNames=张俊, 陈雨艳, 秦震宇, journalName=智慧农业(中英文), refType=null, unstructuredReference=张俊, 陈雨艳, 秦震宇,
等. 改进DeepLab v3+模型下的梯田遥感提取研究[J].
智慧农业(中英文),
2024,
6(3): 46-57., articleTitle=改进DeepLab v3+模型下的梯田遥感提取研究, refAbstract=null), Reference(id=1317882767921234689, tenantId=1146029695717560320, journalId=1281212937352253451, articleId=1317882353167479691, doi=null, pmid=null, pmcid=null, year=2024, volume=6, issue=3, pageStart=46, pageEnd=57, url=null, language=null, rfNumber=[16], rfOrder=24, authorNames=ZHANG J, CHEN Y Y, QIN Z Y, journalName=Smart Agriculture, refType=null, unstructuredReference=
ZHANG J,
CHEN Y Y,
QIN Z Y,
et al. Remote sensing extraction method of terraced fields based on improved DeepLab v3+[J].
Smart Agriculture,
2024,
6(3): 46-57., articleTitle=null, refAbstract=null), Reference(id=1317882767984149250, tenantId=1146029695717560320, journalId=1281212937352253451, articleId=1317882353167479691, doi=null, pmid=null, pmcid=null, year=2021, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=[17], rfOrder=25, authorNames=HU E J, SHEN Y L, WALLIS P, journalName=null, refType=null, unstructuredReference=
HU E J,
SHEN Y L,
WALLIS P,
et al. LoRA: Low-rank adaptation of large language models[EB/OL]. arXiv:
2021., articleTitle=LoRA: Low-rank adaptation of large language models, refAbstract=null), Reference(id=1317882768206447363, tenantId=1146029695717560320, journalId=1281212937352253451, articleId=1317882353167479691, doi=null, pmid=null, pmcid=null, year=2023, volume=null, issue=null, pageStart=3274, pageEnd=3287, url=null, language=null, rfNumber=[18], rfOrder=26, authorNames=VALIPOUR M, REZAGHOLIZADEH M, KOBYZEV I, journalName=null, refType=null, unstructuredReference=
VALIPOUR M,
REZAGHOLIZADEH M,
KOBYZEV I,
et al. DyLoRA: Parameter-efficient tuning of pre-trained models using dynamic search-free low-rank adaptation[C]// Proceedings of the 17th Conference of the European Chapter of the Association for Computational Linguistics. Stroudsburg, PA, USA: ACL,
2023: 3274-3287., articleTitle=DyLoRA: Parameter-efficient tuning of pre-trained models using dynamic search-free low-rank adaptation, refAbstract=null), Reference(id=1317882768281944837, tenantId=1146029695717560320, journalId=1281212937352253451, articleId=1317882353167479691, doi=null, pmid=null, pmcid=null, year=2023, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=[19], rfOrder=27, authorNames=ZHANG Q R, CHEN M S, BUKHARIN A, journalName=null, refType=null, unstructuredReference=
ZHANG Q R,
CHEN M S,
BUKHARIN A,
et al. AdaLoRA: Adaptive budget allocation for parameter-efficient fine-tuning[EB/OL]. arXiv:
2023., articleTitle=AdaLoRA: Adaptive budget allocation for parameter-efficient fine-tuning, refAbstract=null), Reference(id=1317882768353248006, tenantId=1146029695717560320, journalId=1281212937352253451, articleId=1317882353167479691, doi=null, pmid=null, pmcid=null, year=1905, volume=0041, issue=null, pageStart=4, pageEnd=null, url=null, language=null, rfNumber=[20], rfOrder=28, authorNames=KORNBLITH S, NOROUZI M, LEE H, journalName=arXiv, refType=null, unstructuredReference=
KORNBLITH S,
NOROUZI M,
LEE H,
et al. Similarity of neural network representations revisited[EB/OL].
arXiv:
1905.
0041 4, 2019., articleTitle=Similarity of neural network representations revisited, refAbstract=null), Reference(id=1317882768634266375, tenantId=1146029695717560320, journalId=1281212937352253451, articleId=1317882353167479691, doi=null, pmid=null, pmcid=null, year=2022, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=[21], rfOrder=29, authorNames=DAVARI M, HOROI S, NATIK A, journalName=null, refType=null, unstructuredReference=
DAVARI M,
HOROI S,
NATIK A,
et al. Reliability of CKA as a similarity measure in deep learning[EB/OL]. arXiv:
2022., articleTitle=Reliability of CKA as a similarity measure in deep learning, refAbstract=null), Reference(id=1317882768718152456, tenantId=1146029695717560320, journalId=1281212937352253451, articleId=1317882353167479691, doi=null, pmid=null, pmcid=null, year=2018, volume=null, issue=null, pageStart=4510, pageEnd=4520, url=null, language=null, rfNumber=[22], rfOrder=30, authorNames=SANDLER M, HOWARD A, ZHU M L, journalName=null, refType=null, unstructuredReference=
SANDLER M,
HOWARD A,
ZHU M L,
et al. MobileNetV2: Inverted residuals and linear bottlenecks[C]// 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition. Piscataway, New Jersey, USA: IEEE,
2018: 4510-4520., articleTitle=MobileNetV2: Inverted residuals and linear bottlenecks, refAbstract=null), Reference(id=1317882769053696777, tenantId=1146029695717560320, journalId=1281212937352253451, articleId=1317882353167479691, doi=null, pmid=null, pmcid=null, year=2020, volume=null, issue=null, pageStart=10425, pageEnd=10433, url=null, language=null, rfNumber=[23], rfOrder=31, authorNames=RADOSAVOVIC I, KOSARAJU R P, GIRSHICK R, journalName=null, refType=null, unstructuredReference=
RADOSAVOVIC I,
KOSARAJU R P,
GIRSHICK R,
et al. Designing network design spaces[C]// 2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR). Piscataway, New Jersey, USA: IEEE,
2020: 10425-10433., articleTitle=Designing network design spaces, refAbstract=null)], funds=[Fund(id=1317882760715420385, tenantId=1146029695717560320, journalId=1281212937352253451, articleId=1317882353167479691, awardId=32402495, language=EN, fundingSource=National Natural Science Foundation of China(32402495), fundOrder=null, country=null), Fund(id=1317882762355393250, tenantId=1146029695717560320, journalId=1281212937352253451, articleId=1317882353167479691, awardId=32402495, language=CN, fundingSource=国家自然科学基金(32402495), fundOrder=null, country=null), Fund(id=1317882762472833763, tenantId=1146029695717560320, journalId=1281212937352253451, articleId=1317882353167479691, awardId=202410593001S, language=EN, fundingSource=2024 Autonomous Region-Level Student Innovation Training Program Project(202410593001S), fundOrder=null, country=null), Fund(id=1317882762753852133, tenantId=1146029695717560320, journalId=1281212937352253451, articleId=1317882353167479691, awardId=202410593001S, language=CN, fundingSource=2024年自治区级大学生创新训练计划(202410593001S), fundOrder=null, country=null)], companyList=[AuthorCompany(id=1317882750472929937, tenantId=1146029695717560320, journalId=1281212937352253451, articleId=1317882353167479691, xref=1., ext=[AuthorCompanyExt(id=1317882750481318546, tenantId=1146029695717560320, journalId=1281212937352253451, articleId=1317882353167479691, companyId=1317882750472929937, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=
1.School of Computer, Electronics and Information, Guangxi University, Nanning 530000, China), AuthorCompanyExt(id=1317882750577787539, tenantId=1146029695717560320, journalId=1281212937352253451, articleId=1317882353167479691, companyId=1317882750472929937, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=
1.广西大学 计算机与电子信息学院,广西 南宁 530000,中国)]), AuthorCompany(id=1317882750644896404, tenantId=1146029695717560320, journalId=1281212937352253451, articleId=1317882353167479691, xref=2., ext=[AuthorCompanyExt(id=1317882750657479317, tenantId=1146029695717560320, journalId=1281212937352253451, articleId=1317882353167479691, companyId=1317882750644896404, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=
2.State Key Laboratory for Conservation and Utilization of Subtropical Agrobiological Resources, Nanning 530000, China), AuthorCompanyExt(id=1317882750665867926, tenantId=1146029695717560320, journalId=1281212937352253451, articleId=1317882353167479691, companyId=1317882750644896404, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=
2.亚热带农业保护与利用国家重点实验室,广西 南宁 530000,中国)]), AuthorCompany(id=1317882750728782487, tenantId=1146029695717560320, journalId=1281212937352253451, articleId=1317882353167479691, xref=3., ext=[AuthorCompanyExt(id=1317882750741365400, tenantId=1146029695717560320, journalId=1281212937352253451, articleId=1317882353167479691, companyId=1317882750728782487, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=
3.JJR Science and Technology Group Co. , Ltd. , Nanning 530000, China), AuthorCompanyExt(id=1317882750745559705, tenantId=1146029695717560320, journalId=1281212937352253451, articleId=1317882353167479691, companyId=1317882750728782487, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=
3.捷佳润科技集团股份有限公司,广西 南宁 530000,中国)])], figs=[ArticleFig(id=1317882755829056198, tenantId=1146029695717560320, journalId=1281212937352253451, articleId=1317882353167479691, language=EN, label=Fig. 1, caption=
Sample images from the plant disease dataset, figureFileSmall=QN7g6eRXCq9+Gb78Ke/DkQ==, figureFileBig=lQfQfxEDBmhb2s/P7tDkGw==, tableContent=null), ArticleFig(id=1317882755900359367, tenantId=1146029695717560320, journalId=1281212937352253451, articleId=1317882353167479691, language=CN, label=图1, caption=
部分植物病害数据集, figureFileSmall=QN7g6eRXCq9+Gb78Ke/DkQ==, figureFileBig=lQfQfxEDBmhb2s/P7tDkGw==, tableContent=null), ArticleFig(id=1317882756114268872, tenantId=1146029695717560320, journalId=1281212937352253451, articleId=1317882353167479691, language=EN, label=Fig. 2, caption=
The overall framework of dynamic rank allocation (DRA), figureFileSmall=EKsBgVq8UXzjF6jvup2Ndw==, figureFileBig=ZLoW29XVgq/b1SRseVaSdQ==, tableContent=null), ArticleFig(id=1317882756172989129, tenantId=1146029695717560320, journalId=1281212937352253451, articleId=1317882353167479691, language=CN, label=图2, caption=
动态秩分配算法整体框架图, figureFileSmall=EKsBgVq8UXzjF6jvup2Ndw==, figureFileBig=ZLoW29XVgq/b1SRseVaSdQ==, tableContent=null), ArticleFig(id=1317882756235903690, tenantId=1146029695717560320, journalId=1281212937352253451, articleId=1317882353167479691, language=EN, label=Fig. 3, caption=
Changes in layer-wise rank allocation tables of RegNetY and MobileNetV2 models during training, figureFileSmall=hA6H57GdK19zkxtKJGdbuw==, figureFileBig=KzH3TIYDcgdWI6LzQP1oKw==, tableContent=null), ArticleFig(id=1317882757863293643, tenantId=1146029695717560320, journalId=1281212937352253451, articleId=1317882353167479691, language=CN, label=图3, caption=
RegNetY和MobileNetV2模型层级秩分配表在训练过程中的变化a. RegNetY模型 b. MobileNetV2模型
, figureFileSmall=hA6H57GdK19zkxtKJGdbuw==, figureFileBig=KzH3TIYDcgdWI6LzQP1oKw==, tableContent=null), ArticleFig(id=1317882757938791117, tenantId=1146029695717560320, journalId=1281212937352253451, articleId=1317882353167479691, language=EN, label=Fig. 4, caption=
Comparison of validation accuracy for AlexNet and RegNetY models under DRA and LoRA fine-tuning, figureFileSmall=HUm8Oy2VMp0OAfr8W85FQg==, figureFileBig=QCc7lAPVXJxBTjPRYhH75Q==, tableContent=null), ArticleFig(id=1317882758026871502, tenantId=1146029695717560320, journalId=1281212937352253451, articleId=1317882353167479691, language=CN, label=图4, caption=
AlexNet和RegNetY模型DRA与LoRA微调验证集准确率, figureFileSmall=HUm8Oy2VMp0OAfr8W85FQg==, figureFileBig=QCc7lAPVXJxBTjPRYhH75Q==, tableContent=null), ArticleFig(id=1317882758093980367, tenantId=1146029695717560320, journalId=1281212937352253451, articleId=1317882353167479691, language=EN, label=Fig. 5, caption=
Comparison of validation set accuracy peaks and training time among DRA, AdaLoRA, and DyLoRA methods, figureFileSmall=aceFdhsEQUBPSdwZBT7ylw==, figureFileBig=ggQEHxKEzLGQSocHVlxXCQ==, tableContent=null), ArticleFig(id=1317882758156894928, tenantId=1146029695717560320, journalId=1281212937352253451, articleId=1317882353167479691, language=CN, label=图5, caption=
DRA、AdaLoRA和DyLoRA方法的验证集准确率峰值与训练时间对比, figureFileSmall=aceFdhsEQUBPSdwZBT7ylw==, figureFileBig=ggQEHxKEzLGQSocHVlxXCQ==, tableContent=null), ArticleFig(id=1317882758228198097, tenantId=1146029695717560320, journalId=1281212937352253451, articleId=1317882353167479691, language=EN, label=null, caption=null, figureFileSmall=null, figureFileBig=null, tableContent=
| 算法: RRA |
|---|
Input: Pre-trained model , dataset . Predefined: max rank , min rank , total epochs , calibration epochs , Importance weights , Bias weights , . # Initialization the LoRA parameters (, alpha, dropout) for each target layer.lora_config ← ComputeInitialRanks(, ) by formula (1)~(14) # Initialize the LoRA low-rank matrices according to lora_config, where follows a random Gaussian distribution and is initialized as a zero matrix. 0 Initialize optimizer for LoRA parameters # Initialize trigger threshold # Training Loop for epoch = 1 to do Update by minimizing loss on dataset . CalculateAvgParamChange(, ) by formula (17,18) If == then . # Set initial threshold. else if and then # Trigger rank update // Merge LoRA weights for each target layer for each target layer l in M do
end for lora_config′ ← ComputeInitialRanks() # Initialize the new matrices and according to lora_config′. 0 Re-initialize optimizer for new LoRA parameters # Update threshold end if end for return |
), ArticleFig(id=1317882758312084178, tenantId=1146029695717560320, journalId=1281212937352253451, articleId=1317882353167479691, language=CN, label=null, caption=null, figureFileSmall=null, figureFileBig=null, tableContent=
| 算法: RRA |
|---|
Input: Pre-trained model , dataset . Predefined: max rank , min rank , total epochs , calibration epochs , Importance weights , Bias weights , . # Initialization the LoRA parameters (, alpha, dropout) for each target layer.lora_config ← ComputeInitialRanks(, ) by formula (1)~(14) # Initialize the LoRA low-rank matrices according to lora_config, where follows a random Gaussian distribution and is initialized as a zero matrix. 0 Initialize optimizer for LoRA parameters # Initialize trigger threshold # Training Loop for epoch = 1 to do Update by minimizing loss on dataset . CalculateAvgParamChange(, ) by formula (17,18) If == then . # Set initial threshold. else if and then # Trigger rank update // Merge LoRA weights for each target layer for each target layer l in M do
end for lora_config′ ← ComputeInitialRanks() # Initialize the new matrices and according to lora_config′. 0 Re-initialize optimizer for new LoRA parameters # Update threshold end if end for return |
), ArticleFig(id=1317882758387581651, tenantId=1146029695717560320, journalId=1281212937352253451, articleId=1317882353167479691, language=EN, label=Table 1, caption=
Comparison of the accuracy and the training parameters of AlexNet, MobileNetV2, RegNetY and ConvNeXt on each dataset
, figureFileSmall=null, figureFileBig=null, tableContent=
| 模型 | 小麦病害数据集 | 植物病害数据集 |
|---|
| Full fine-tuning | DRA | Full fine-tuning | DRA |
|---|
| mAP/% | Parm/M | mAP/% | Avg_Parm/M | mAP/% | Parm/M | mAP/% | Avg_Parm/M |
|---|
| AlexNet | 81.75 | 217.62 | 80.09 | 0.90 | 90.41 | 217.62 | 89.42 | 0.66 |
| MobileNetV2 | 77.02 | 8.54 | 75.80 | 0.21 | 87.04 | 8.54 | 86.21 | 0.25 |
| RegNetY | 82.07 | 74.67 | 80.67 | 2.66 | 90.83 | 74.67 | 90.96 | 2.10 |
| ConvNeXt | 74.90 | 1 290 | 72.13 | 16.17 | 86.89 | 1 290 | 85.12 | 12.55 |
), ArticleFig(id=1317882758454690516, tenantId=1146029695717560320, journalId=1281212937352253451, articleId=1317882353167479691, language=CN, label=表1, caption=
AlexNet、MobileNetV2、RegNetY和ConvNeXt模型在各数据集上的准确率与训练参数对比
, figureFileSmall=null, figureFileBig=null, tableContent=
| 模型 | 小麦病害数据集 | 植物病害数据集 |
|---|
| Full fine-tuning | DRA | Full fine-tuning | DRA |
|---|
| mAP/% | Parm/M | mAP/% | Avg_Parm/M | mAP/% | Parm/M | mAP/% | Avg_Parm/M |
|---|
| AlexNet | 81.75 | 217.62 | 80.09 | 0.90 | 90.41 | 217.62 | 89.42 | 0.66 |
| MobileNetV2 | 77.02 | 8.54 | 75.80 | 0.21 | 87.04 | 8.54 | 86.21 | 0.25 |
| RegNetY | 82.07 | 74.67 | 80.67 | 2.66 | 90.83 | 74.67 | 90.96 | 2.10 |
| ConvNeXt | 74.90 | 1 290 | 72.13 | 16.17 | 86.89 | 1 290 | 85.12 | 12.55 |
), ArticleFig(id=1317882758517605077, tenantId=1146029695717560320, journalId=1281212937352253451, articleId=1317882353167479691, language=EN, label=Table 2, caption=
Comparison of trainable parameter sizes during fine-tuning of AlexNet, MobileNetV2, RegNetY and ConvNeXt models with the RRA strategy on various datasets
, figureFileSmall=null, figureFileBig=null, tableContent=
| 数据集 | 模型 | Epoch | Parm/M | Epoch | Parm/M | Epoch | Parm/M | Epoch | Parm/M | Epoch | Parm/M |
|---|
| 小麦病害数据集 | AlexNet | T0 | 1.35 | T7 | 1.01 | T17 | 0.76 | T29 | 0.72 | / | / |
| MobileNetV2 | T0 | 0.33 | T7 | 0.20 | T13 | 0.19 | T21 | 0.19 | T30 | 0.19 |
| RegNetY | T0 | 3.56 | T7 | 2.58 | T15 | 2.54 | T26 | 2.44 | T39 | 2.31 |
| ConvNeXt | T0 | 19.62 | T7 | 18.21 | T14 | 16.71 | T24 | 13.80 | T35 | 13.73 |
| 植物病害数据集 | AlexNet | T0 | 1.38 | T7 | 0.62 | T19 | 0.49 | / | / | / | / |
| MobileNetV2 | T0 | 0.27 | T7 | 0.26 | T15 | 0.25 | T25 | 0.23 | T37 | 0.23 |
| RegNetY | T0 | 3.64 | T7 | 1.83 | / | / | / | / | / | / |
| ConvNeXt | T0 | 18.34 | T7 | 13.58 | T14 | 13.47 | T22 | 10.45 | T33 | 9.26 |
), ArticleFig(id=1317882758601491158, tenantId=1146029695717560320, journalId=1281212937352253451, articleId=1317882353167479691, language=CN, label=表2, caption=
AlexNet、MobileNetV2、RegNetY和ConvNeXt模型应用RRA策略在各数据集微调过程中的训练参数大小对比
, figureFileSmall=null, figureFileBig=null, tableContent=
| 数据集 | 模型 | Epoch | Parm/M | Epoch | Parm/M | Epoch | Parm/M | Epoch | Parm/M | Epoch | Parm/M |
|---|
| 小麦病害数据集 | AlexNet | T0 | 1.35 | T7 | 1.01 | T17 | 0.76 | T29 | 0.72 | / | / |
| MobileNetV2 | T0 | 0.33 | T7 | 0.20 | T13 | 0.19 | T21 | 0.19 | T30 | 0.19 |
| RegNetY | T0 | 3.56 | T7 | 2.58 | T15 | 2.54 | T26 | 2.44 | T39 | 2.31 |
| ConvNeXt | T0 | 19.62 | T7 | 18.21 | T14 | 16.71 | T24 | 13.80 | T35 | 13.73 |
| 植物病害数据集 | AlexNet | T0 | 1.38 | T7 | 0.62 | T19 | 0.49 | / | / | / | / |
| MobileNetV2 | T0 | 0.27 | T7 | 0.26 | T15 | 0.25 | T25 | 0.23 | T37 | 0.23 |
| RegNetY | T0 | 3.64 | T7 | 1.83 | / | / | / | / | / | / |
| ConvNeXt | T0 | 18.34 | T7 | 13.58 | T14 | 13.47 | T22 | 10.45 | T33 | 9.26 |
), ArticleFig(id=1317882758693765847, tenantId=1146029695717560320, journalId=1281212937352253451, articleId=1317882353167479691, language=EN, label=Table 3, caption=
Performance comparison of DRA and LoRA fine-tuning on various models
, figureFileSmall=null, figureFileBig=null, tableContent=
| 模型 | LoRA(R=4) | LoRA(R=8) | LoRA(R=16) | DRA |
|---|
| mAP/% | Parm/M | mAP/% | Parm/M | mAP/% | Parm/M | mAP/% | Parm/M |
|---|
| AlexNet | 77.64 | 0.55 | 79.93 | 1.11 | 79.71 | 2.21 | 80.09 | 0.90 |
| MobileNetV2 | 75.54 | 0.41 | 75.51 | 0.83 | 75.40 | 1.66 | 75.80 | 0.21 |
| RegNetY | 79.67 | 1.34 | 80.79 | 2.69 | 80.86 | 5.37 | 80.91 | 2.66 |
| ConvNeXt | 70.26 | 6.29 | 71.42 | 12.57 | 72.64 | 25.14 | 72.13 | 16.17 |
), ArticleFig(id=1317882758786040536, tenantId=1146029695717560320, journalId=1281212937352253451, articleId=1317882353167479691, language=CN, label=表3, caption=
AlexNet、MobileNetV2、RegNetY和ConvNeXt模型DRA与LoRA微调性能对比
, figureFileSmall=null, figureFileBig=null, tableContent=
| 模型 | LoRA(R=4) | LoRA(R=8) | LoRA(R=16) | DRA |
|---|
| mAP/% | Parm/M | mAP/% | Parm/M | mAP/% | Parm/M | mAP/% | Parm/M |
|---|
| AlexNet | 77.64 | 0.55 | 79.93 | 1.11 | 79.71 | 2.21 | 80.09 | 0.90 |
| MobileNetV2 | 75.54 | 0.41 | 75.51 | 0.83 | 75.40 | 1.66 | 75.80 | 0.21 |
| RegNetY | 79.67 | 1.34 | 80.79 | 2.69 | 80.86 | 5.37 | 80.91 | 2.66 |
| ConvNeXt | 70.26 | 6.29 | 71.42 | 12.57 | 72.64 | 25.14 | 72.13 | 16.17 |
), ArticleFig(id=1317882758966395609, tenantId=1146029695717560320, journalId=1281212937352253451, articleId=1317882353167479691, language=EN, label=Table 4, caption=
Accuracy of ablation experiments on AlexNet, MobileNetV2, RegNetY and ConvNeXt models under different and weight combinations on wheat disease dataset
, figureFileSmall=null, figureFileBig=null, tableContent=
| 模型 | = 0, = 1 | = 0.15, = 0.85 | = 0.20, = 0.80 | = 0.25, = 0.75 | = 1, = 0 |
|---|
| AlexNet | 76.87 | 80.09 | 79.46 | 78.55 | 77.63 |
| MobileNetV2 | 74.58 | 75.80 | 76.79 | 76.94 | 74.73 |
| RegNetY | 73.76 | 80.67 | 80.38 | 80.99 | 79.24 |
| ConvNeXt | 65.65 | 72.13 | 71.26 | 70.81 | 69.66 |
), ArticleFig(id=1317882759037698778, tenantId=1146029695717560320, journalId=1281212937352253451, articleId=1317882353167479691, language=CN, label=表4, caption=
不同和权重组合下AlexNet、MobileNetV2、RegNetY和ConvNeXt模型在小麦病害数据集上的消融实验准确率 ( %)
, figureFileSmall=null, figureFileBig=null, tableContent=
| 模型 | = 0, = 1 | = 0.15, = 0.85 | = 0.20, = 0.80 | = 0.25, = 0.75 | = 1, = 0 |
|---|
| AlexNet | 76.87 | 80.09 | 79.46 | 78.55 | 77.63 |
| MobileNetV2 | 74.58 | 75.80 | 76.79 | 76.94 | 74.73 |
| RegNetY | 73.76 | 80.67 | 80.38 | 80.99 | 79.24 |
| ConvNeXt | 65.65 | 72.13 | 71.26 | 70.81 | 69.66 |
), ArticleFig(id=1317882759104807643, tenantId=1146029695717560320, journalId=1281212937352253451, articleId=1317882353167479691, language=EN, label=Table 5, caption=
Trainable parameter count of AlexNet, MobileNetV2, RegNetY and ConvNeXt models under differentand weight combinations on wheat disease dataset
, figureFileSmall=null, figureFileBig=null, tableContent=
| 模型 | = 0, = 1 | = 0.15, = 0.85 | = 0.20, = 0.80 | = 0.25, = 0.75 | = 1, = 0 |
|---|
| AlexNet | 0.70 | 0.90 | 0.95 | 1.17 | 1.89 |
| MobileNetV2 | 0.15 | 0.18 | 0.18 | 0.21 | 1.16 |
| RegNetY | 0.67 | 2.66 | 2.88 | 3.42 | 4.30 |
| ConvNeXt | 10.26 | 16.17 | 17.53 | 19.3 | 24.12 |
), ArticleFig(id=1317882759410991836, tenantId=1146029695717560320, journalId=1281212937352253451, articleId=1317882353167479691, language=CN, label=表5, caption=
不同和权重组合下AlexNet、MobileNetV2、RegNetY和ConvNeXt模型在小麦病害数据集上的可训练参数量 ( M)
, figureFileSmall=null, figureFileBig=null, tableContent=
| 模型 | = 0, = 1 | = 0.15, = 0.85 | = 0.20, = 0.80 | = 0.25, = 0.75 | = 1, = 0 |
|---|
| AlexNet | 0.70 | 0.90 | 0.95 | 1.17 | 1.89 |
| MobileNetV2 | 0.15 | 0.18 | 0.18 | 0.21 | 1.16 |
| RegNetY | 0.67 | 2.66 | 2.88 | 3.42 | 4.30 |
| ConvNeXt | 10.26 | 16.17 | 17.53 | 19.3 | 24.12 |
), ArticleFig(id=1317882759511655133, tenantId=1146029695717560320, journalId=1281212937352253451, articleId=1317882353167479691, language=EN, label=Table 6, caption=
Accuracy of the ConvNeXt model with DRA method in robustness test under four data perturbations
, figureFileSmall=null, figureFileBig=null, tableContent=
| 数据集 | DRA | Gaussian noise | Random crop | Color jitter | Random rotation |
|---|
| 植物病害数据集 | 85.12 | 83.02 | 83.98 | 79.10 | 80.01 |
| 小麦病害数据集 | 72.13 | 68.93 | 67.80 | 70.42 | 67.73 |
), ArticleFig(id=1317882759813645022, tenantId=1146029695717560320, journalId=1281212937352253451, articleId=1317882353167479691, language=CN, label=表6, caption=
ConvNeXt模型应用DRA方法在4种数据干扰下进行鲁棒性测试的准确率 ( %)
, figureFileSmall=null, figureFileBig=null, tableContent=
| 数据集 | DRA | Gaussian noise | Random crop | Color jitter | Random rotation |
|---|
| 植物病害数据集 | 85.12 | 83.02 | 83.98 | 79.10 | 80.01 |
| 小麦病害数据集 | 72.13 | 68.93 | 67.80 | 70.42 | 67.73 |
), ArticleFig(id=1317882759910114015, tenantId=1146029695717560320, journalId=1281212937352253451, articleId=1317882353167479691, language=EN, label=Table 7, caption=
Comparison of fine-tuning performance and parameter budget among DRA, AdaLoRA, and DyLoRA methods
, figureFileSmall=null, figureFileBig=null, tableContent=
| 模型 | 方法 | 小麦病害数据集 | 植物病害数据集 |
|---|
| Avg_Parm/M | mAP/% | Avg_Parm/M | mAP/% |
|---|
| AlexNet | DRA | 0.90 | 80.09 | 0.66 | 89.42 |
| AdaLoRA | 0.90 | 79.95 | 0.66 | 88.65 |
| DyLoRA | 1.35 | 80.16 | 0.99 | 89.25 |
| DyLoRA | 0.90 | 78.71 | 0.66 | 88.03 |
| DyLoRA | 0.45 | 76.88 | 0.33 | 86.57 |
| RegNetY | DRA | 2.66 | 80.67 | 2.10 | 90.96 |
| AdaLoRA | 2.66 | 80.55 | 2.10 | 90.41 |
| DyLoRA | 3.99 | 80.81 | 3.15 | 90.75 |
| DyLoRA | 2.66 | 79.83 | 2.10 | 90.02 |
| DyLoRA | 1.33 | 78.52 | 1.05 | 88.94 |
), ArticleFig(id=1317882760245658336, tenantId=1146029695717560320, journalId=1281212937352253451, articleId=1317882353167479691, language=CN, label=表7, caption=
DRA、AdaLoRA和DyLoRA方法的微调性能与参数预算对比
, figureFileSmall=null, figureFileBig=null, tableContent=
| 模型 | 方法 | 小麦病害数据集 | 植物病害数据集 |
|---|
| Avg_Parm/M | mAP/% | Avg_Parm/M | mAP/% |
|---|
| AlexNet | DRA | 0.90 | 80.09 | 0.66 | 89.42 |
| AdaLoRA | 0.90 | 79.95 | 0.66 | 88.65 |
| DyLoRA | 1.35 | 80.16 | 0.99 | 89.25 |
| DyLoRA | 0.90 | 78.71 | 0.66 | 88.03 |
| DyLoRA | 0.45 | 76.88 | 0.33 | 86.57 |
| RegNetY | DRA | 2.66 | 80.67 | 2.10 | 90.96 |
| AdaLoRA | 2.66 | 80.55 | 2.10 | 90.41 |
| DyLoRA | 3.99 | 80.81 | 3.15 | 90.75 |
| DyLoRA | 2.66 | 79.83 | 2.10 | 90.02 |
| DyLoRA | 1.33 | 78.52 | 1.05 | 88.94 |
)], attaches=null, journal=Journal(id=1281211802411339782, delFlag=0, nameCn=智慧农业(中英文), nameEn=Smart Agriculture, nameHistory1=null, nameHistory2=null, issn=2096-8094, eissn=2097-485X, cn=10-1681/S, coden=ZNZHD7, periodic=1, language=CN, oaType=null, ccby=null, superviseOffice=null, ownerOffice=null, pubOffice=null, editorOffice=null, officeType=null, aims=null, clcCode=null, officeProv=null, officeCity=null, officeAddr=null, officeZip=null, officeEmail=null, officePhone=null, editDirector=null, officeDirector=null, officeDirectorPhone=null, officeStaffNum=null, officeEmpNum=null, coverPicUrl=3ySu/Wank9ycKptCQ9bxHg==, journalPrice=null, startedYear=null, abbrevIsoEn=Smart Agriculture, journalRemark=null, publicationField=null, createdTime=1783394297461, updatedTime=1788949493025, createdBy=18614031015, updatedBy=13041195026, firstLetterCn=Z, firstLetterEn=Z, subjectCode=Agriculture, subjectName=null, subjectCodeEn=Agriculture, subjectNameEn=null, picCn=3ySu/Wank9ycKptCQ9bxHg==, picEn=6aR2ETcq7WHPaEneIXN2aw==, jcr=null, cjcr=null, exts=[JournalExt(id=1304511981487157723, language=CN, name=智慧农业(中英文), nameHistory1=null, nameHistory2=null, managedBy=, sponsoredBy=, publishedBy=, editorOffice=, officeProv=null, officeCity=null, officeAddr=, officeZip=, editDirector=, officeDirector=null, officePhone=null, coverPicUrl=null, journalRemark=, submitArticleUrl=null, websiteUrl=, createdTime=1788949493047, updatedTime=1788949493047, createdBy=13041195026, updatedBy=13041195026, submissionGuidelinesUrl=, submissionAuthorUrl=http://www.smartag.net.cn/Journalx_zhny/authorLogOn.action, submissionEditorUrl=http://www.smartag.net.cn/Journalx_zhny/editorLogOn.action, submissionReviewUrl=http://www.smartag.net.cn/Journalx_zhny/expertLogOn.action, submissionCeEditorUrl=, submissionAeEditorUrl=, option={"copyright":""}), JournalExt(id=1304511981529100764, language=EN, name=Smart Agriculture, nameHistory1=null, nameHistory2=null, managedBy=, sponsoredBy=, publishedBy=, editorOffice=, officeProv=null, officeCity=null, officeAddr=, officeZip=, editDirector=, officeDirector=null, officePhone=null, coverPicUrl=null, journalRemark=, submitArticleUrl=null, websiteUrl=, createdTime=1788949493057, updatedTime=1788949493057, createdBy=13041195026, updatedBy=13041195026, submissionGuidelinesUrl=, submissionAuthorUrl=http://www.smartag.net.cn/Journalx_zhny/authorLogOn.action, submissionEditorUrl=http://www.smartag.net.cn/Journalx_zhny/editorLogOn.action, submissionReviewUrl=http://www.smartag.net.cn/Journalx_zhny/expertLogOn.action, submissionCeEditorUrl=, submissionAeEditorUrl=, option={"copyright":""})], databaseList=null, tenantJournalId=1281212937352253451, websiteList=[Website(id=1281216327134188517, webName=null, webTitle=null, webDomain=null, webCopyrigh=null, webIpcNo=null, seoTitle=null, seoKeywords=null, seoDescription=null, tenantJournalId=null, journalId=1281212937352253451, journalNameCn=null, journalNameEn=null, grayFlag=null, tenantId=1146029695717560320, platformId=null, journalGroupId=null, journalGroupNameCn=null, journalGroupNameEn=null, type=1, domain=https://castjournals.cast.org.cn/joweb/zhny/CN, language=CN, createTime=1783395376237, createBy=18614031015, updateTime=1783395541424, updateBy=18614031015, name=智慧农业(中英文)-中文, tplId=1146099689490845704, title=智慧农业(中英文), delFlag=0, indexPage=/home, props=[WebsiteProps(id=1281217159418319048, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1281216327134188517, code=articleTextType, value=kx, createTime=1783395574669, updateTime=1783395574669, creator=18614031015, updator=18614031015), WebsiteProps(id=1281217159388958917, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1281216327134188517, code=banner, value=null, createTime=1783395574662, updateTime=1783395574662, creator=18614031015, updator=18614031015), WebsiteProps(id=1281217159443484875, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1281216327134188517, code=grayFlag, value=0, createTime=1783395574675, updateTime=1783395574675, creator=18614031015, updator=18614031015), WebsiteProps(id=1281217159355404484, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1281216327134188517, code=logo, value=https://castjournals.cast.org.cn/joweb/zhny/CN/file/pic?fileId=RbfWW55ASXpJGAr1dytAUA==, createTime=1783395574654, updateTime=1783395574654, creator=18614031015, updator=18614031015), WebsiteProps(id=1281217159456067789, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1281216327134188517, code=minRunFlag, value=0, createTime=1783395574678, updateTime=1783395574678, creator=18614031015, updator=18614031015), WebsiteProps(id=1281217159414124743, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1281216327134188517, code=picServerUrl, value=https://castjournals.cast.org.cn/joweb/zhny/CN/file/pic, createTime=1783395574668, updateTime=1783395574668, creator=18614031015, updator=18614031015), WebsiteProps(id=1281217159451873484, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1281216327134188517, code=silenceFlag, value=0, createTime=1783395574677, updateTime=1783395574677, creator=18614031015, updator=18614031015), WebsiteProps(id=1281217159401541830, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1281216327134188517, code=staticResourcePath, value=https://castjournals.cast.org.cn/joweb/cast_kjdb_cn_619/, createTime=1783395574665, updateTime=1783395574665, creator=18614031015, updator=18614031015), WebsiteProps(id=1281217159426707657, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1281216327134188517, code=themeColor, value=null, createTime=1783395574671, updateTime=1783395574671, creator=18614031015, updator=18614031015), WebsiteProps(id=1281217159435096266, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1281216327134188517, code=themeStyle, value=null, createTime=1783395574673, updateTime=1783395574673, creator=18614031015, updator=18614031015)]), Website(id=1281216327880774635, webName=null, webTitle=null, webDomain=null, webCopyrigh=null, webIpcNo=null, seoTitle=null, seoKeywords=null, seoDescription=null, tenantJournalId=null, journalId=1281212937352253451, journalNameCn=null, journalNameEn=null, grayFlag=null, tenantId=1146029695717560320, platformId=null, journalGroupId=null, journalGroupNameCn=null, journalGroupNameEn=null, type=1, domain=https://castjournals.cast.org.cn/joweb/zhny/EN, language=EN, createTime=1783395376415, createBy=18614031015, updateTime=1783395537123, updateBy=18614031015, name=智慧农业(中英文)-英文, tplId=1146101810881728533, title=Smart Agriculture, delFlag=0, indexPage=/home, props=[WebsiteProps(id=1281217117114568888, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1281216327880774635, code=articleTextType, value=kx, createTime=1783395564583, updateTime=1783395564583, creator=18614031015, updator=18614031015), WebsiteProps(id=1281217117089403061, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1281216327880774635, code=banner, value=null, createTime=1783395564577, updateTime=1783395564577, creator=18614031015, updator=18614031015), WebsiteProps(id=1281217117135540411, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1281216327880774635, code=grayFlag, value=0, createTime=1783395564588, updateTime=1783395564588, creator=18614031015, updator=18614031015), WebsiteProps(id=1281217117064237236, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1281216327880774635, code=logo, value=https://castjournals.cast.org.cn/joweb/zhny/EN/file/pic?fileId=RbfWW55ASXpJGAr1dytAUA==, createTime=1783395564571, updateTime=1783395564571, creator=18614031015, updator=18614031015), WebsiteProps(id=1281217117148123325, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1281216327880774635, code=minRunFlag, value=0, createTime=1783395564591, updateTime=1783395564591, creator=18614031015, updator=18614031015), WebsiteProps(id=1281217117106180279, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1281216327880774635, code=picServerUrl, value=https://castjournals.cast.org.cn/joweb/zhny/EN/file/pic, createTime=1783395564581, updateTime=1783395564581, creator=18614031015, updator=18614031015), WebsiteProps(id=1281217117139734716, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1281216327880774635, code=silenceFlag, value=0, createTime=1783395564589, updateTime=1783395564589, creator=18614031015, updator=18614031015), WebsiteProps(id=1281217117097791670, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1281216327880774635, code=staticResourcePath, value=https://castjournals.cast.org.cn/joweb/cast_kjdb_en_623/, createTime=1783395564579, updateTime=1783395564579, creator=18614031015, updator=18614031015), WebsiteProps(id=1281217117122957497, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1281216327880774635, code=themeColor, value=null, createTime=1783395564585, updateTime=1783395564585, creator=18614031015, updator=18614031015), WebsiteProps(id=1281217117127151802, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1281216327880774635, code=themeStyle, value=null, createTime=1783395564586, updateTime=1783395564586, creator=18614031015, updator=18614031015)])], journalTitle=智慧农业(中英文), weixinUrl=null, journalUrl=https://www.smartag.net.cn/, iacademicId=null, status=1, seqNo=null, journalTitleEn=Smart Agriculture, journalPhotoCn=3ySu/Wank9ycKptCQ9bxHg==, journalPhotoEn=6aR2ETcq7WHPaEneIXN2aw==, journalFirstLetter=Z, journalRecommend=null, journalNew=null, journalCollection=null, jcrJf=null, cjcrJf=null, jcrJfStr=null, cjcrJfStr=null, submissionFirstDecision=null, sciSubjectClassification=null, casSubjectClassification=null, citeScore=null, totalCitationFrequency=null, icpCode=null, psCode=null, advertisingLicenseCode=null, copyrightInformation=null, country=null, option=, provinceCode=null, provinceName=null, collectFlag=false, interPubPlatform=, interPubPlatformUrl=null), detailUrlCn=https://castjournals.cast.org.cn/joweb/zhny/CN/10.12133/j.smartag.SA202504003, detailUrlEn=https://castjournals.cast.org.cn/joweb/zhny/EN/10.12133/j.smartag.SA202504003, pdfUrlCn=https://castjournals.cast.org.cn/joweb/zhny/CN/PDF/10.12133/j.smartag.SA202504003, pdfUrlEn=https://castjournals.cast.org.cn/joweb/zhny/EN/PDF/10.12133/j.smartag.SA202504003, aliStartDate=0, aliEndDate=0, collectionFlag=false, citedCount=null, citedUrl=null, previewStatus=0, delFlag=0, hasFullText=1, orderTime=1769702400000, fullTextJson=null, articleText=null, reference=null)