Article(id=1251856527338975832, tenantId=1146029695717560320, journalId=1251234268282663017, issueId=1251856520619700745, articleNumber=null, orderNo=null, doi=10.3969/j.issn.1003-3106.2025.11.006, pmid=null, cstr=null, oa=null, hot=null, price=null, onlineType=0, articleFormat=0, articleType=null, articleTypeStr=null, receivedDate=1751299200000, receivedDateStr=2025-07-01, revisedDate=null, revisedDateStr=null, acceptedDate=null, acceptedDateStr=null, onlineDate=1776395454486, onlineDateStr=2026-04-17, pubDate=1762272000000, pubDateStr=2025-11-05, doiRegisterDate=null, doiRegisterDateStr=null, onlineIssueDate=1776395454486, onlineIssueDateStr=2026-04-17, onlineJustAcceptDate=null, onlineJustAcceptDateStr=null, onlineFirstDate=null, onlineFirstDateStr=null, sourceXml=null, magXml=null, createTime=1776395454486, creator=13701087609, updateTime=1776395454486, updator=13701087609, issue=Issue{id=1251856520619700745, tenantId=1146029695717560320, journalId=1251234268282663017, year='2025', volume='55', issue='11', pageStart='2131', pageEnd='2324', issueExtLink='null', onlineDate='null', pubDate='null', beforeIssueId=null, nextIssueId=null, price=null, status=1, issueComplete=1, articleOrder=1, issueType=1, specialIssue=null, createTime=1776395452885, creator=13701087609, updateTime=1776395571911, updator=13701087609, preIssue=null, nextIssue=null, ext={EN=IssueExt(id=1251857019939013255, tenantId=1146029695717560320, journalId=1251234268282663017, issueId=1251856520619700745, language=EN, specialIssueTitle=, coverIllustrator=null, specialIssueEditor=, specialIssueAbout=), CN=IssueExt(id=1251857019939013256, tenantId=1146029695717560320, journalId=1251234268282663017, issueId=1251856520619700745, language=CN, specialIssueTitle=, coverIllustrator=null, specialIssueEditor=, specialIssueAbout=)}, issueFiles=null}, startPage=2184, endPage=2194, ext={EN=ArticleExt(id=1251856527582245473, articleId=1251856527338975832, tenantId=1146029695717560320, journalId=1251234268282663017, language=EN, title=Research on Algorithm for Weed Recognition in Beet Fields Based on Improved YOLOv11, columnId=1251856523492798993, journalTitle=Radio Engineering, columnName=Signal and Information Processing, runingTitle=null, highlight=null, articleAbstract=

An algorithm for weed recognition in beet fields based on improved YOLOv11 model is proposed to address the problems of low efficiency, low accuracy, and missed detection of small targets in complex real-world scenarios. The PoolFormer module and AKConv module are introduced into the backbone network to enhance the model's ability to capture global semantic information to improve detection accuracy, enhancing the detection performance in low resolution images and small objects. The AKConv module improves the feature extraction ability of the model for beets and weeds with irregular growth patterns by dynamically adjusting the convolution kernel parameters and shapes, while the PoolFormer module can effectively segment the edge features of beets and weeds that cover each other. Secondly, the High-level Screening Feature Pyramid Network (HS-FPN) module is added to the head network to enhance the efficiency of multi-scale fusion and improve the feature extraction efficiency and speed of beets and weeds during the seedling stage. Through experiments, it is found that the improved YOLOv11 model achieves increases of 6.9%, 7.8%, 7.9%, and 7.8% in precision, recall, mAP@0.5 and mAP@0.5: 0.95, respectively, compared to the original model. The results show that this algorithm has achieved significant improvement in weed recognition in beet fields, providing a more feasible solution for detecting weeds in beet fields in complex scenarios.

, correspAuthors=Qiang LIU, authorNote=null, correspAuthorsNote=null, copyrightStatement=null, 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, authorCompany=null, fund=null, authors=null, authorsList=Zijian ZHOU, Qiang LIU), CN=ArticleExt(id=1251856532942566214, articleId=1251856527338975832, tenantId=1146029695717560320, journalId=1251234268282663017, language=CN, title=基于改进的YOLOv11甜菜田间杂草识别算法研究, columnId=1251856523803177493, journalTitle=无线电工程, columnName=信号与信息处理, runingTitle=null, highlight=null, articleAbstract=

针对复杂真实场景下甜菜田间杂草识别效率低、精确率低、小目标漏检等问题,提出一种基于YOLOv11模型改进的甜菜田间杂草识别算法。在主干网络引入PoolFormer模块和AKConv模块增强模型对全局语义信息的捕捉能力来提升检测精度,提高了在低分辨率图像和小物体中的检测效果。AKConv模块通过动态调整卷积核参数和形状,提升模型对生长形态不规则的甜菜与杂草的特征提取能力;PoolFormer模块可以很好地将相互遮盖的甜菜与杂草的边缘特征进行分割。在头部网络加入高级筛选-特征融合金字塔网络(High-level Screening Feature Pyramid Network,HS-FPN)模块增强多尺度融合效率,提升幼苗期甜菜与杂草的特征提取效率和速度。通过实验得出,改进YOLOv11模型的精确率、召回率、mAP@0.5、mAP@0.5:0.95相较于改进前分别提升了6.9%、7.8%、7.9%、7.8%。结果表明,在甜菜田间杂草识别上取得了显著的提升效果,为复杂场景中检测甜菜田间杂草提供了一个更可行的解决方案。

, correspAuthors=刘强, authorNote=null, correspAuthorsNote=
刘强 男,(1974—),硕士,教授,硕士生导师。主要研究方向:农业信息化。
, copyrightStatement=null, copyrightOwner=null, extLink=null, articleAbsUrl=null, sourceXml=gDbdNqlPXE+6EIxurAvO4A==, magXml=oMvMNQT1nWehJrFZXJG3Fg==, pdfUrl=null, pdf=yL5bDvBsbiRkaCqNAdnBfg==, pdfFileSize=18759477, pdfExtLink=null, richHtmlUrl=null, mobilePdfUrl=null, reviewReport=null, pdfFirstPage=null, abstractGraph=cmfTHhBBkp9nEOZcpyVHKw==, abstractGraphContent=null, abstractVideo=null, citation=null, cebUrl=null, magXmlContent=U1+FZghNfyZ23mLu0+QFIA==, mapNumber=null, authorCompany=null, fund=null, authors=

周子健 男,(1998—),硕士研究生。主要研究方向:深度学习、图像识别。

, authorsList=周子健, 刘强)}, authors=[Author(id=1251856533244556122, tenantId=1146029695717560320, journalId=1251234268282663017, articleId=1251856527338975832, orderNo=0, 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=1251856533324247906, tenantId=1146029695717560320, journalId=1251234268282663017, articleId=1251856527338975832, authorId=1251856533244556122, language=EN, stringName=Zijian ZHOU, firstName=Zijian, middleName=null, lastName=ZHOU, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=null, address=College of Information Science and Technology, Gansu Agricultural University, Lanzhou 730070, China, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null), CN=AuthorExt(id=1251856533424911208, tenantId=1146029695717560320, journalId=1251234268282663017, articleId=1251856527338975832, authorId=1251856533244556122, language=CN, stringName=周子健, firstName=null, middleName=null, lastName=null, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=null, address=甘肃农业大学 信息科学技术学院,甘肃 兰州 730070, bio={"content":"

周子健 男,(1998—),硕士研究生。主要研究方向:深度学习、图像识别。

"}, bioImg=null, bioContent=

周子健 男,(1998—),硕士研究生。主要研究方向:深度学习、图像识别。

, aboutCorrespAuthor=null)}, companyList=[AuthorCompany(id=1251856533148087121, tenantId=1146029695717560320, journalId=1251234268282663017, articleId=1251856527338975832, xref=null, ext=[AuthorCompanyExt(id=1251856533156475730, tenantId=1146029695717560320, journalId=1251234268282663017, articleId=1251856527338975832, companyId=1251856533148087121, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=College of Information Science and Technology, Gansu Agricultural University, Lanzhou 730070, China), AuthorCompanyExt(id=1251856533164864339, tenantId=1146029695717560320, journalId=1251234268282663017, articleId=1251856527338975832, companyId=1251856533148087121, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=甘肃农业大学 信息科学技术学院,甘肃 兰州 730070)])]), Author(id=1251856533487825775, tenantId=1146029695717560320, journalId=1251234268282663017, articleId=1251856527338975832, orderNo=1, 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=1, authorType=1, ext={EN=AuthorExt(id=1251856533563323256, tenantId=1146029695717560320, journalId=1251234268282663017, articleId=1251856527338975832, authorId=1251856533487825775, language=EN, stringName=Qiang LIU, firstName=Qiang, middleName=null, lastName=LIU, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=*, address=College of Information Science and Technology, Gansu Agricultural University, Lanzhou 730070, China, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null), CN=AuthorExt(id=1251856533659792255, tenantId=1146029695717560320, journalId=1251234268282663017, articleId=1251856527338975832, authorId=1251856533487825775, language=CN, stringName=刘强, firstName=null, middleName=null, lastName=null, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=*, address=甘肃农业大学 信息科学技术学院,甘肃 兰州 730070, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null)}, companyList=[AuthorCompany(id=1251856533148087121, tenantId=1146029695717560320, journalId=1251234268282663017, articleId=1251856527338975832, xref=null, ext=[AuthorCompanyExt(id=1251856533156475730, tenantId=1146029695717560320, journalId=1251234268282663017, articleId=1251856527338975832, companyId=1251856533148087121, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=College of Information Science and Technology, Gansu Agricultural University, Lanzhou 730070, China), AuthorCompanyExt(id=1251856533164864339, tenantId=1146029695717560320, journalId=1251234268282663017, articleId=1251856527338975832, companyId=1251856533148087121, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=甘肃农业大学 信息科学技术学院,甘肃 兰州 730070)])])], keywords=[Keyword(id=1251856533760455555, tenantId=1146029695717560320, journalId=1251234268282663017, articleId=1251856527338975832, language=EN, orderNo=1, keyword=YOLOv11), Keyword(id=1251856533848535947, tenantId=1146029695717560320, journalId=1251234268282663017, articleId=1251856527338975832, language=EN, orderNo=2, keyword=beet), Keyword(id=1251856533961782163, tenantId=1146029695717560320, journalId=1251234268282663017, articleId=1251856527338975832, language=EN, orderNo=3, keyword=weed identification), Keyword(id=1251856534058251160, tenantId=1146029695717560320, journalId=1251234268282663017, articleId=1251856527338975832, language=EN, orderNo=4, keyword=edge feature segmentation), Keyword(id=1251856534129554336, tenantId=1146029695717560320, journalId=1251234268282663017, articleId=1251856527338975832, language=EN, orderNo=5, keyword=small target detection), Keyword(id=1251856534205051815, tenantId=1146029695717560320, journalId=1251234268282663017, articleId=1251856527338975832, language=EN, orderNo=6, keyword=deep learning), Keyword(id=1251856534309909423, tenantId=1146029695717560320, journalId=1251234268282663017, articleId=1251856527338975832, language=CN, orderNo=1, keyword=YOLOv11), Keyword(id=1251856534414767035, tenantId=1146029695717560320, journalId=1251234268282663017, articleId=1251856527338975832, language=CN, orderNo=2, keyword=甜菜), Keyword(id=1251856534523818947, tenantId=1146029695717560320, journalId=1251234268282663017, articleId=1251856527338975832, language=CN, orderNo=3, keyword=杂草识别), Keyword(id=1251856536063128523, tenantId=1146029695717560320, journalId=1251234268282663017, articleId=1251856527338975832, language=CN, orderNo=4, keyword=边缘特征分割), Keyword(id=1251856536142820308, tenantId=1146029695717560320, journalId=1251234268282663017, articleId=1251856527338975832, language=CN, orderNo=5, keyword=小目标检测), Keyword(id=1251856536251872219, tenantId=1146029695717560320, journalId=1251234268282663017, articleId=1251856527338975832, language=CN, orderNo=6, keyword=深度学习)], refs=[Reference(id=1251856538919448718, tenantId=1146029695717560320, journalId=1251234268282663017, articleId=1251856527338975832, doi=null, pmid=null, pmcid=null, year=2025, volume=52, issue=8, pageStart=214, pageEnd=221, url=null, language=null, rfNumber=[1], rfOrder=0, authorNames=沈涛, 张秀再, 许岱, journalName=计算机科学, refType=null, unstructuredReference=沈涛,张秀再,许岱.改进RT-DETR的遥感图像小目标检测算法[J].计算机科学,2025,52(8):214-221., articleTitle=改进RT-DETR的遥感图像小目标检测算法, refAbstract=null), Reference(id=1251856540509089939, tenantId=1146029695717560320, journalId=1251234268282663017, articleId=1251856527338975832, doi=null, pmid=null, pmcid=null, year=2023, volume=44, issue=1, pageStart=34, pageEnd=42, url=null, language=null, rfNumber=[2], rfOrder=1, authorNames=胡炼, 刘海龙, 何杰, journalName=华南农业大学学报, refType=null, unstructuredReference=胡炼,刘海龙,何杰,.智能除草机器人研究现状与展望[J].华南农业大学学报,2023,44(1):34-42., articleTitle=智能除草机器人研究现状与展望, refAbstract=null), Reference(id=1251856540668473497, tenantId=1146029695717560320, journalId=1251234268282663017, articleId=1251856527338975832, doi=null, pmid=null, pmcid=null, year=2020, volume=51, issue=增刊2, pageStart=323, pageEnd=334, url=null, language=null, rfNumber=[3], rfOrder=2, authorNames=袁洪波, 赵努东, 程曼, journalName=农业机械学报, refType=null, unstructuredReference=袁洪波,赵努东,程曼.基于图像处理的田间杂草识别研究进展与展望[J].农业机械学报,2020,51(增刊2):323-334., articleTitle=基于图像处理的田间杂草识别研究进展与展望, refAbstract=null), Reference(id=1251856540781719714, tenantId=1146029695717560320, journalId=1251234268282663017, articleId=1251856527338975832, doi=null, pmid=null, pmcid=null, year=2025, volume=null, issue=2, pageStart=1, pageEnd=7, url=null, language=null, rfNumber=[4], rfOrder=3, authorNames=陈雪, 谭峰, journalName=农机使用与维修, refType=null, unstructuredReference=陈雪,谭峰.基于改进YOLOv5的轻量级大豆田间杂草识别方法研究[J].农机使用与维修,2025(2):1-7., articleTitle=基于改进YOLOv5的轻量级大豆田间杂草识别方法研究, refAbstract=null), Reference(id=1251856540890771622, tenantId=1146029695717560320, journalId=1251234268282663017, articleId=1251856527338975832, doi=null, pmid=null, pmcid=null, year=2025-06-15, volume=null, issue=null, pageStart=1, pageEnd=13, url=null, language=null, rfNumber=[5], rfOrder=4, authorNames=文韬, 王天一, 黄诗锐, journalName=计算机工程与科学, refType=null, unstructuredReference=文韬,王天一,黄诗锐,.基于改进YOLOv8的农作物与藜草检测算法:MES-YOLO[J/OL].计算机工程与科学,1-13[2025-06-15]. https://link.cnki.net/urlid/43.1258.tp.20241011.1309.004., articleTitle=基于改进YOLOv8的农作物与藜草检测算法:MES-YOLO, refAbstract=null), Reference(id=1251856541075321004, tenantId=1146029695717560320, journalId=1251234268282663017, articleId=1251856527338975832, doi=null, pmid=null, pmcid=null, year=2025-05-15, volume=null, issue=null, pageStart=1, pageEnd=10, url=null, language=null, rfNumber=[6], rfOrder=5, authorNames=杜希望, 李霞, 唐腾飞, journalName=天津理工大学学报, refType=null, unstructuredReference=杜希望,李霞,唐腾飞,.基于改进YOLOv8n的双模态杂草检测方法[J/OL].天津理工大学学报,1-10[2025-05-15]. https://link.cnki.net/urlid/12.1374.N.20241029.1920.024., articleTitle=基于改进YOLOv8n的双模态杂草检测方法, refAbstract=null), Reference(id=1251856541167595697, tenantId=1146029695717560320, journalId=1251234268282663017, articleId=1251856527338975832, doi=null, pmid=null, pmcid=null, year=2024, volume=55, issue=1, pageStart=212, pageEnd=222, url=null, language=null, rfNumber=[7], rfOrder=6, authorNames=冀汶莉, 刘洲, 邢海花, journalName=农业机械学报, refType=null, unstructuredReference=冀汶莉,刘洲,邢海花.基于YOLO v5的农田杂草识别轻量化方法研究[J].农业机械学报,2024,55(1):212-222., articleTitle=基于YOLO v5的农田杂草识别轻量化方法研究, refAbstract=null), Reference(id=1251856541280841910, tenantId=1146029695717560320, journalId=1251234268282663017, articleId=1251856527338975832, doi=null, pmid=null, pmcid=null, year=2019, volume=35, issue=20, pageStart=202, pageEnd=209, url=null, language=null, rfNumber=[8], rfOrder=7, authorNames=彭明霞, 夏俊芳, 彭辉, journalName=农业工程学报, refType=null, unstructuredReference=彭明霞,夏俊芳,彭辉.融合FPN的Faster R-CNN复杂背景下棉田杂草高效识别方法[J].农业工程学报, 2019,35(20):202-209., articleTitle=融合FPN的Faster R-CNN复杂背景下棉田杂草高效识别方法, refAbstract=null), Reference(id=1251856541343756475, tenantId=1146029695717560320, journalId=1251234268282663017, articleId=1251856527338975832, doi=null, pmid=null, pmcid=null, year=2024, volume=47, issue=24, pageStart=60, pageEnd=67, url=null, language=null, rfNumber=[9], rfOrder=8, authorNames=杨明轩, 陈琳, journalName=现代电子技术, refType=null, unstructuredReference=杨明轩,陈琳.改进YOLOv5的棉田杂草检测[J].现代电子技术,2024,47(24):60-67., articleTitle=改进YOLOv5的棉田杂草检测, refAbstract=null), Reference(id=1251856541410865344, tenantId=1146029695717560320, journalId=1251234268282663017, articleId=1251856527338975832, doi=null, pmid=null, pmcid=null, year=2025, volume=46, issue=2, pageStart=237, pageEnd=244, url=null, language=null, rfNumber=[10], rfOrder=9, authorNames=罗柳茗, 李岩舟, 石美琦, journalName=中国农机化学报, refType=null, unstructuredReference=罗柳茗,李岩舟,石美琦,.基于YOLOv8n的甘蔗杂草识别方法[J].中国农机化学报,2025,46(2): 237-244., articleTitle=基于YOLOv8n的甘蔗杂草识别方法, refAbstract=null), Reference(id=1251856541482168514, tenantId=1146029695717560320, journalId=1251234268282663017, articleId=1251856527338975832, doi=null, pmid=null, pmcid=null, year=2024, volume=63, issue=8, pageStart=17, pageEnd=22, url=null, language=null, rfNumber=[11], rfOrder=10, authorNames=林宗缪, 马超, 胡冬, journalName=湖北农业科学, refType=null, unstructuredReference=林宗缪,马超,胡冬.基于改进YOLOv8卷积神经网络的稻田苗期杂草检测方法[J].湖北农业科学,2024, 63(8):17-22., articleTitle=基于改进YOLOv8卷积神经网络的稻田苗期杂草检测方法, refAbstract=null), Reference(id=1251856541574443207, tenantId=1146029695717560320, journalId=1251234268282663017, articleId=1251856527338975832, doi=null, pmid=null, pmcid=null, year=2023, volume=14, issue=null, pageStart=1183277, pageEnd=null, url=null, language=null, rfNumber=[12], rfOrder=11, authorNames=DARBYSHIRE M, SALAZAR-GOMEZ A, GAO J F, journalName=Frontiers in Plant Science, refType=null, unstructuredReference=DARBYSHIRE M,SALAZAR-GOMEZ A,GAO J F,et al. Towards Practical Object Detection for Weed Spraying in Precision Agriculture[J]. Frontiers in Plant Science, 2023,14:1183277., articleTitle=Towards Practical Object Detection for Weed Spraying in Precision Agriculture, refAbstract=null), Reference(id=1251856541687689421, tenantId=1146029695717560320, journalId=1251234268282663017, articleId=1251856527338975832, doi=null, pmid=null, pmcid=null, year=2025-05-10, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=[13], rfOrder=12, authorNames=KHANAM R, HUSSAIN M, journalName=null, refType=null, unstructuredReference=KHANAM R,HUSSAIN M. YOLOv11: An Overview of the Key Architectural Enhancements[EB/OL]. (2024-10-23)[2025-05-10]. https://arxiv.org/abs/2410.17725., articleTitle=YOLOv11: An Overview of the Key Architectural Enhancements, refAbstract=null), Reference(id=1251856541805129938, tenantId=1146029695717560320, journalId=1251234268282663017, articleId=1251856527338975832, doi=null, pmid=null, pmcid=null, year=2024, volume=14, issue=17, pageStart=7541, pageEnd=null, url=null, language=null, rfNumber=[14], rfOrder=13, authorNames=ZHANG R T, ZHU D J, WU C, journalName=Applied Sciences, refType=null, unstructuredReference=ZHANG R T,ZHU D J,WU C,et al. Sensitive Information Detection Based on Deep Learning Models[J]. Applied Sciences,2024,14(17):7541., articleTitle=Sensitive Information Detection Based on Deep Learning Models, refAbstract=null), Reference(id=1251856541884821722, tenantId=1146029695717560320, journalId=1251234268282663017, articleId=1251856527338975832, doi=null, pmid=null, pmcid=null, year=2025-05-11, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=[15], rfOrder=14, authorNames=HIDAYATULLAH P, SYAKRANI N, SHOLAHUDDIN M R, journalName=null, refType=null, unstructuredReference=HIDAYATULLAH P,SYAKRANI N,SHOLAHUDDIN M R,et al. YOLOv8 to YOLO11: A Comprehensive Architecture In-depth Comparative Review[EB/OL]. (2025-01-23)[2025-05-11]. https://arxiv.org/abs/2501.13400., articleTitle=YOLOv8 to YOLO11: A Comprehensive Architecture In-depth Comparative Review, refAbstract=null), Reference(id=1251856541998067936, tenantId=1146029695717560320, journalId=1251234268282663017, articleId=1251856527338975832, doi=null, pmid=null, pmcid=null, year=2025, volume=55, issue=8, pageStart=1703, pageEnd=1709, url=null, language=null, rfNumber=[16], rfOrder=15, authorNames=马艺嘉, 卢睿智, 付可欣, journalName=无线电工程, refType=null, unstructuredReference=马艺嘉,卢睿智,付可欣,.基于MBL-YOLOv8n模型的光伏缺陷检测[J].无线电工程,2025,55(8):1703-1709., articleTitle=基于MBL-YOLOv8n模型的光伏缺陷检测, refAbstract=null), Reference(id=1251856542115508454, tenantId=1146029695717560320, journalId=1251234268282663017, articleId=1251856527338975832, doi=null, pmid=null, pmcid=null, year=2025, volume=55, issue=8, pageStart=1571, pageEnd=1579, url=null, language=null, rfNumber=[17], rfOrder=16, authorNames=李璐, 陈清江, journalName=无线电工程, refType=null, unstructuredReference=李璐,陈清江.基于超分辨率和YOLO的红外小目标检测[J].无线电工程,2025,55(8):1571-1579., articleTitle=基于超分辨率和YOLO的红外小目标检测, refAbstract=null), Reference(id=1251856542195200235, tenantId=1146029695717560320, journalId=1251234268282663017, articleId=1251856527338975832, doi=null, pmid=null, pmcid=null, year=2020, volume=58, issue=null, pageStart=11093, pageEnd=null, url=null, language=null, rfNumber=[18], rfOrder=17, authorNames=QIN J H, PAN W Y, XIANG X Y, journalName=Ecological Informatics, refType=null, unstructuredReference=QIN J H,PAN W Y,XIANG X Y,et al. A Biological Image Classification Method Based on Improved CNN[J]. Ecological Informatics,2020,58:11093., articleTitle=A Biological Image Classification Method Based on Improved CNN, refAbstract=null), Reference(id=1251856542274892016, tenantId=1146029695717560320, journalId=1251234268282663017, articleId=1251856527338975832, doi=null, pmid=null, pmcid=null, year=2023, volume=125, issue=null, pageStart=106729, pageEnd=null, url=null, language=null, rfNumber=[19], rfOrder=18, authorNames=YANG E Q, ZHOU W J, QIAN X H, journalName=Engineering Applications of Artificial Intelligence, refType=null, unstructuredReference=YANG E Q,ZHOU W J,QIAN X H,et al. DRNet: Dual-stage Refinement Network with Boundary Inference for RGB-D Semantic Segmentation of Indoor Scenes[J]. Engineering Applications of Artificial Intelligence, 2023, 125:106729., articleTitle=DRNet: Dual-stage Refinement Network with Boundary Inference for RGB-D Semantic Segmentation of Indoor Scenes, refAbstract=null), Reference(id=1251856542358778101, tenantId=1146029695717560320, journalId=1251234268282663017, articleId=1251856527338975832, doi=null, pmid=null, pmcid=null, year=2023, volume=135, issue=null, pageStart=104697, pageEnd=null, url=null, language=null, rfNumber=[20], rfOrder=19, authorNames=WANG X, HE N, HONG C, journalName=Image and Vision Computing, refType=null, unstructuredReference=WANG X, HE N, HONG C, et al. Improved YOLOX-X Based UAV Aerial Photography Object Detection Algorithm[J]. Image and Vision Computing,2023,135:104697., articleTitle=Improved YOLOX-X Based UAV Aerial Photography Object Detection Algorithm, refAbstract=null), Reference(id=1251856542455247093, tenantId=1146029695717560320, journalId=1251234268282663017, articleId=1251856527338975832, doi=null, pmid=null, pmcid=null, year=2025-05-11, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=[21], rfOrder=20, authorNames=ZHU X Z, HU H, LIN S, journalName=null, refType=null, unstructuredReference=ZHU X Z,HU H,LIN S,et al. Deformable ConvNets v2:More Deformable,Better Results[EB/OL]. (2018-11-27)[2025-05-11]. https://arxiv.org/abs/1811.11168., articleTitle=Deformable ConvNets v2:More Deformable,Better Results, refAbstract=null), Reference(id=1251856542551716091, tenantId=1146029695717560320, journalId=1251234268282663017, articleId=1251856527338975832, doi=null, pmid=null, pmcid=null, year=2025-05-11, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=[22], rfOrder=21, authorNames=ZHANG X, SONG Y Z, SONG T T, journalName=null, refType=null, unstructuredReference=ZHANG X,SONG Y Z,SONG T T,et al. AKConv: Convolutional Kernel with Arbitrary Sampled Shapes and Arbitrary Number of Parameters[EB/OL]. (2023-11-20)[2025-05-11]. https://arxiv.org/abs/2311.11587v1., articleTitle=AKConv: Convolutional Kernel with Arbitrary Sampled Shapes and Arbitrary Number of Parameters, refAbstract=null), Reference(id=1251856542639796480, tenantId=1146029695717560320, journalId=1251234268282663017, articleId=1251856527338975832, doi=null, pmid=null, pmcid=null, year=2024, volume=14, issue=10, pageStart=2355, pageEnd=null, url=null, language=null, rfNumber=[23], rfOrder=22, authorNames=JIA Z Y, ZHANG M, YUAN C, journalName=Agronomy, refType=null, unstructuredReference=JIA Z Y, ZHANG M, YUAN C, et al. ADL-YOLOv8: A Field Crop Weed Detection Model Based on Improved YOLOv8[J]. Agronomy,2024,14(10):2355., articleTitle=ADL-YOLOv8: A Field Crop Weed Detection Model Based on Improved YOLOv8, refAbstract=null), Reference(id=1251856542740459780, tenantId=1146029695717560320, journalId=1251234268282663017, articleId=1251856527338975832, doi=null, pmid=null, pmcid=null, year=2024, volume=13, issue=15, pageStart=3001, pageEnd=null, url=null, language=null, rfNumber=[24], rfOrder=23, authorNames=SUN H, TAN C H, PANG S, journalName=Electronics, refType=null, unstructuredReference=SUN H,TAN C H,PANG S,et al. RA-YOLOv8: An Improved YOLOv8 Seal Text Detection Method[J]. Electronics,2024,13(15):3001., articleTitle=RA-YOLOv8: An Improved YOLOv8 Seal Text Detection Method, refAbstract=null), Reference(id=1251856542857900296, tenantId=1146029695717560320, journalId=1251234268282663017, articleId=1251856527338975832, doi=null, pmid=null, pmcid=null, year=2025-05-11, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=[25], rfOrder=24, authorNames=CHEN Y F, ZHANG C Y, CHEN B, journalName=null, refType=null, unstructuredReference=CHEN Y F,ZHANG C Y,CHEN B,et al. Accurate Leukocyte Detection Based on Deformable-DETR and Multilevel Feature Fusion for Aiding Diagnosis of Blood Diseases[EB/OL]. (2024-01-01)[2025-05-11]. https://arxiv.org/abs/2401.00926., articleTitle=Accurate Leukocyte Detection Based on Deformable-DETR and Multilevel Feature Fusion for Aiding Diagnosis of Blood Diseases, refAbstract=null), Reference(id=1251856543004700943, tenantId=1146029695717560320, journalId=1251234268282663017, articleId=1251856527338975832, doi=null, pmid=null, pmcid=null, year=2024, volume=13, issue=11, pageStart=1710, pageEnd=null, url=null, language=null, rfNumber=[26], rfOrder=25, authorNames=LIU R M, SU W H, journalName=Foods, refType=null, unstructuredReference=LIU R M,SU W H. APHS-YOLO: A Lightweight Model for Real-time Detection and Classification of Stropharia Rugoso-Annulata[J]. Foods,2024,13(11):1710., articleTitle=APHS-YOLO: A Lightweight Model for Real-time Detection and Classification of Stropharia Rugoso-Annulata, refAbstract=null), Reference(id=1251856543092781333, tenantId=1146029695717560320, journalId=1251234268282663017, articleId=1251856527338975832, doi=null, pmid=null, pmcid=null, year=2024, volume=15, issue=null, pageStart=1409544, pageEnd=null, url=null, language=null, rfNumber=[27], rfOrder=26, authorNames=XIN D Y, LI T Q, journalName=Frontiers in Plant Science, refType=null, unstructuredReference=XIN D Y,LI T Q. Revolutionizing Tomato Disease Detection in Complex Environments[J]. Frontiers in Plant Science,2024,15:1409544., articleTitle=Revolutionizing Tomato Disease Detection in Complex Environments, refAbstract=null), Reference(id=1251856543180861721, tenantId=1146029695717560320, journalId=1251234268282663017, articleId=1251856527338975832, doi=null, pmid=null, pmcid=null, year=2023, volume=null, issue=null, pageStart=2195, pageEnd=2200, url=null, language=null, rfNumber=[28], rfOrder=27, authorNames=MA X W, YANG J W, HONG T F, journalName=null, refType=null, unstructuredReference=MA X W,YANG J W,HONG T F,et al. STNet: Spatial and Temporal Feature Fusion Network For Change Detection in Remote Sensing Images[C]//2023 IEEE International Conference on Multimedia and Expo (ICME). Brisbane:IEEE,2023:2195-2200., articleTitle=STNet: Spatial and Temporal Feature Fusion Network For Change Detection in Remote Sensing Images, refAbstract=null), Reference(id=1251856543281525022, tenantId=1146029695717560320, journalId=1251234268282663017, articleId=1251856527338975832, doi=null, pmid=null, pmcid=null, year=2025-05-12, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=[29], rfOrder=28, authorNames=VASWANI A, SHAZEER N, PARMAR N, journalName=null, refType=null, unstructuredReference=VASWANI A,SHAZEER N,PARMAR N,et al. Attention Is All You Need[EB/OL]. (2017-06-12)[2025-05-12]. https://arxiv.org/abs/1706.03762., articleTitle=Attention Is All You Need, refAbstract=null), Reference(id=1251856543369605411, tenantId=1146029695717560320, journalId=1251234268282663017, articleId=1251856527338975832, doi=null, pmid=null, pmcid=null, year=2025-05-10, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=[30], rfOrder=29, authorNames=YU W H, LUO M, ZHOU P, journalName=null, refType=null, unstructuredReference=YU W H,LUO M,ZHOU P,et al. MetaFormer Is Actually What You Need for Vision[EB/OL]. (2021-11-12)[2025-05-10]. https://arxiv.org/abs/2111.11418., articleTitle=MetaFormer Is Actually What You Need for Vision, refAbstract=null), Reference(id=1251856544950858026, tenantId=1146029695717560320, journalId=1251234268282663017, articleId=1251856527338975832, doi=null, pmid=null, pmcid=null, year=2024, volume=31, issue=6, pageStart=66, pageEnd=75, url=null, language=null, rfNumber=[31], rfOrder=30, authorNames=LIAO Y N, HUANG C Y, ZHANG T, journalName=The Journal of China Universities of Posts and Telecommunications, refType=null, unstructuredReference=LIAO Y N,HUANG C Y,ZHANG T,et al. China 05 Bone Age Assessment Based on Deep Learning[J]. The Journal of China Universities of Posts and Telecommunications, 2024,31(6):66-75., articleTitle=China 05 Bone Age Assessment Based on Deep Learning, refAbstract=null)], funds=null, companyList=[AuthorCompany(id=1251856533148087121, tenantId=1146029695717560320, journalId=1251234268282663017, articleId=1251856527338975832, xref=null, ext=[AuthorCompanyExt(id=1251856533156475730, tenantId=1146029695717560320, journalId=1251234268282663017, articleId=1251856527338975832, companyId=1251856533148087121, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=College of Information Science and Technology, Gansu Agricultural University, Lanzhou 730070, China), AuthorCompanyExt(id=1251856533164864339, tenantId=1146029695717560320, journalId=1251234268282663017, articleId=1251856527338975832, companyId=1251856533148087121, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=甘肃农业大学 信息科学技术学院,甘肃 兰州 730070)])], figs=[ArticleFig(id=1251856536373507048, tenantId=1146029695717560320, journalId=1251234268282663017, articleId=1251856527338975832, language=EN, label=Fig.1, caption=YOLOv11 network architecture, figureFileSmall=4LTUFysoetRNAKy80+Cwpw==, figureFileBig=cmfTHhBBkp9nEOZcpyVHKw==, tableContent=null), ArticleFig(id=1251856536469976043, tenantId=1146029695717560320, journalId=1251234268282663017, articleId=1251856527338975832, language=CN, label=图1, caption=YOLOv11网络结构, figureFileSmall=4LTUFysoetRNAKy80+Cwpw==, figureFileBig=cmfTHhBBkp9nEOZcpyVHKw==, tableContent=null), ArticleFig(id=1251856536704857087, tenantId=1146029695717560320, journalId=1251234268282663017, articleId=1251856527338975832, language=EN, label=Fig.2, caption=Improved YOLOv11 network architecture, figureFileSmall=VZ4sU7VcOzjRd3AKjnc6OA==, figureFileBig=Rgb3ZC/ZD0XqSeOrTlo42A==, tableContent=null), ArticleFig(id=1251856536813907974, tenantId=1146029695717560320, journalId=1251234268282663017, articleId=1251856527338975832, language=CN, label=图2, caption=改进的YOLOv11网络结构, figureFileSmall=VZ4sU7VcOzjRd3AKjnc6OA==, figureFileBig=Rgb3ZC/ZD0XqSeOrTlo42A==, tableContent=null), ArticleFig(id=1251856536914571279, tenantId=1146029695717560320, journalId=1251234268282663017, articleId=1251856527338975832, language=EN, label=Fig.3, caption=AKConv network architecture, figureFileSmall=/UNi1topWuJiz8uCOSxMVg==, figureFileBig=7G3yAZwbdJaMeLweWytalA==, tableContent=null), ArticleFig(id=1251856537015234581, tenantId=1146029695717560320, journalId=1251234268282663017, articleId=1251856527338975832, language=CN, label=图3, caption=AKConv网络结构, figureFileSmall=/UNi1topWuJiz8uCOSxMVg==, figureFileBig=7G3yAZwbdJaMeLweWytalA==, tableContent=null), ArticleFig(id=1251856537132675102, tenantId=1146029695717560320, journalId=1251234268282663017, articleId=1251856527338975832, language=EN, label=Fig.4, caption=HS-FPN network architecture, figureFileSmall=6c/LI3wrGiEsubb2qFsERQ==, figureFileBig=vf6swCZPI1hUKUs6bb0y5A==, tableContent=null), ArticleFig(id=1251856537245921318, tenantId=1146029695717560320, journalId=1251234268282663017, articleId=1251856527338975832, language=CN, label=图4, caption=HS-FPN网络结构, figureFileSmall=6c/LI3wrGiEsubb2qFsERQ==, figureFileBig=vf6swCZPI1hUKUs6bb0y5A==, tableContent=null), ArticleFig(id=1251856537334001708, tenantId=1146029695717560320, journalId=1251234268282663017, articleId=1251856527338975832, language=EN, label=Fig.5, caption=SFF network architecture, figureFileSmall=sqrDEj9fNyLD+w5pzlr+Xw==, figureFileBig=PXOAHSywGMPqySF4WHI20w==, tableContent=null), ArticleFig(id=1251856537409499186, tenantId=1146029695717560320, journalId=1251234268282663017, articleId=1251856527338975832, language=CN, label=图5, caption=SFF网络结构, figureFileSmall=sqrDEj9fNyLD+w5pzlr+Xw==, figureFileBig=PXOAHSywGMPqySF4WHI20w==, tableContent=null), ArticleFig(id=1251856537510162487, tenantId=1146029695717560320, journalId=1251234268282663017, articleId=1251856527338975832, language=EN, label=Fig.6, caption=Network architecture of each module, figureFileSmall=ta/r9jnAUCbB0NuuQNFKNg==, figureFileBig=vkrhJe/p8V8G8yQLwLl2XQ==, tableContent=null), ArticleFig(id=1251856537598242880, tenantId=1146029695717560320, journalId=1251234268282663017, articleId=1251856527338975832, language=CN, label=图6, caption=各架构网络结构, figureFileSmall=ta/r9jnAUCbB0NuuQNFKNg==, figureFileBig=vkrhJe/p8V8G8yQLwLl2XQ==, tableContent=null), ArticleFig(id=1251856537703100486, tenantId=1146029695717560320, journalId=1251234268282663017, articleId=1251856527338975832, language=EN, label=Fig.7, caption=Pooling network architecture, figureFileSmall=wTU1wO6K1TZVT5dgU1ygUg==, figureFileBig=tDtom2TE9Z5EKLi1omlc2A==, tableContent=null), ArticleFig(id=1251856537782792267, tenantId=1146029695717560320, journalId=1251234268282663017, articleId=1251856527338975832, language=CN, label=图7, caption=Pooling网络结构, figureFileSmall=wTU1wO6K1TZVT5dgU1ygUg==, figureFileBig=tDtom2TE9Z5EKLi1omlc2A==, tableContent=null), ArticleFig(id=1251856537875066960, tenantId=1146029695717560320, journalId=1251234268282663017, articleId=1251856527338975832, language=EN, label=Fig.8, caption=Comparison of training set loss, figureFileSmall=9InNFo6LM2U9JvMcABw8WA==, figureFileBig=PY+q4+wffeS8gste+iDKEQ==, tableContent=null), ArticleFig(id=1251856537950564441, tenantId=1146029695717560320, journalId=1251234268282663017, articleId=1251856527338975832, language=CN, label=图8, caption=训练集损失, figureFileSmall=9InNFo6LM2U9JvMcABw8WA==, figureFileBig=PY+q4+wffeS8gste+iDKEQ==, tableContent=null), ArticleFig(id=1251856538038644830, tenantId=1146029695717560320, journalId=1251234268282663017, articleId=1251856527338975832, language=EN, label=Fig.9, caption=Comparison of validation set loss, figureFileSmall=sUXZtbNP2fNZhtjzK57Idg==, figureFileBig=Zm5p4O4yHvvp0J9gsAZJgg==, tableContent=null), ArticleFig(id=1251856538147696738, tenantId=1146029695717560320, journalId=1251234268282663017, articleId=1251856527338975832, language=CN, label=图9, caption=验证集损失, figureFileSmall=sUXZtbNP2fNZhtjzK57Idg==, figureFileBig=Zm5p4O4yHvvp0J9gsAZJgg==, tableContent=null), ArticleFig(id=1251856538290303079, tenantId=1146029695717560320, journalId=1251234268282663017, articleId=1251856527338975832, language=EN, label=Fig.10, caption=Example diagram of visual analysis, figureFileSmall=wJ0TXLY/ja7XhSrbaxnJ2g==, figureFileBig=J9tL72G7Pb889HFfbwPpeA==, tableContent=null), ArticleFig(id=1251856538374189165, tenantId=1146029695717560320, journalId=1251234268282663017, articleId=1251856527338975832, language=CN, label=图10, caption=可视化分析示例, figureFileSmall=wJ0TXLY/ja7XhSrbaxnJ2g==, figureFileBig=J9tL72G7Pb889HFfbwPpeA==, tableContent=null), ArticleFig(id=1251856538466463860, tenantId=1146029695717560320, journalId=1251234268282663017, articleId=1251856527338975832, language=EN, label=Tab.1, caption=

Ablation experiment data for differentmodule improvements

, figureFileSmall=null, figureFileBig=null, tableContent=
模型名称PRmAP@0.5mAP@0.5:0.95
AKConv80.073.879.354.1
HS-FPN79.873.679.353.9
PoolFormer80.975.580.955.1
YOLO11-improve81.977.182.356.0
YOLOv1176.771.576.351.8
), ArticleFig(id=1251856538554544253, tenantId=1146029695717560320, journalId=1251234268282663017, articleId=1251856527338975832, language=CN, label=表1, caption=

不同模块改进的消融实验数据

, figureFileSmall=null, figureFileBig=null, tableContent=
模型名称PRmAP@0.5mAP@0.5:0.95
AKConv80.073.879.354.1
HS-FPN79.873.679.353.9
PoolFormer80.975.580.955.1
YOLO11-improve81.977.182.356.0
YOLOv1176.771.576.351.8
), ArticleFig(id=1251856538701344896, tenantId=1146029695717560320, journalId=1251234268282663017, articleId=1251856527338975832, language=EN, label=Tab.2, caption=

Comparison experiment data for different models

, figureFileSmall=null, figureFileBig=null, tableContent=
模型名称PRmAP@0.5mAP@0.5:0.95
YOLOv973.767.873.250.1
YOLOv1075.470.175.151.4
YOLOv1176.671.576.351.9
YOLOv11-improve81.977.182.356.0
), ArticleFig(id=1251856538781036676, tenantId=1146029695717560320, journalId=1251234268282663017, articleId=1251856527338975832, language=CN, label=表2, caption=

不同模型对比实验数据

, figureFileSmall=null, figureFileBig=null, tableContent=
模型名称PRmAP@0.5mAP@0.5:0.95
YOLOv973.767.873.250.1
YOLOv1075.470.175.151.4
YOLOv1176.671.576.351.9
YOLOv11-improve81.977.182.356.0
)], attaches=null, journal=Journal(id=1251231494547484770, delFlag=0, nameCn=无线电工程, nameEn=Radio Engineering, nameHistory1=null, nameHistory2=null, issn=1003-3106, eissn=, cn=13-1097/TN, coden=null, periodic=月刊, language=CN, oaType=1, 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=, officePhone=, editDirector=null, officeDirector=null, officeDirectorPhone=null, officeStaffNum=null, officeEmpNum=null, coverPicUrl=g0Lcfmfo7oUYDLtQTsujLQ==, journalPrice=null, startedYear=null, abbrevIsoEn=Radio Engineering, journalRemark=null, publicationField=null, createdTime=1776246435060, updatedTime=1776252047215, createdBy=18614031015, updatedBy=13701087609, firstLetterCn=R, firstLetterEn=R, subjectCode=Engineering, subjectName=工程, subjectCodeEn=Engineering, subjectNameEn=null, picCn=g0Lcfmfo7oUYDLtQTsujLQ==, picEn=fe/NRE6hx/Z5ZHweFTw2gA==, jcr=null, cjcr=null, exts=[JournalExt(id=1251255033736676338, language=CN, name=无线电工程, 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=1776252047237, updatedTime=1776252047237, createdBy=13701087609, updatedBy=13701087609, submissionGuidelinesUrl=, submissionAuthorUrl=https://wxdg.cbpt.cnki.net/index.aspx?t=1, submissionEditorUrl=https://wxdg.cbpt.cnki.net/index.aspx?t=3, submissionReviewUrl=https://wxdg.cbpt.cnki.net/index.aspx?t=2, submissionCeEditorUrl=, submissionAeEditorUrl=, option={"copyright":""}), JournalExt(id=1251255033787007987, language=EN, name=Radio Engineering, 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=1776252047249, updatedTime=1776252047249, createdBy=13701087609, updatedBy=13701087609, submissionGuidelinesUrl=, submissionAuthorUrl=https://wxdg.cbpt.cnki.net/index.aspx?t=1, submissionEditorUrl=https://wxdg.cbpt.cnki.net/index.aspx?t=3, submissionReviewUrl=https://wxdg.cbpt.cnki.net/index.aspx?t=2, submissionCeEditorUrl=, submissionAeEditorUrl=, option={"copyright":""})], databaseList=null, tenantJournalId=1251234268282663017, websiteList=[Website(id=1251257283494232110, webName=null, webTitle=null, webDomain=null, webCopyrigh=null, webIpcNo=null, seoTitle=null, seoKeywords=null, seoDescription=null, tenantJournalId=null, journalId=1251234268282663017, 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/wxdgc/CN, language=CN, createTime=1776252583625, createBy=18614031015, updateTime=1776253601640, updateBy=18614031015, name=无线电工程-中文, tplId=1146099689490845704, title=无线电工程, delFlag=0, indexPage=/home, props=[WebsiteProps(id=1251261682236997912, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1251257283494232110, code=articleTextType, value=kx, createTime=1776253632363, updateTime=1776253632363, creator=18614031015, updator=18614031015), WebsiteProps(id=1251261682199249173, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1251257283494232110, code=banner, value=null, createTime=1776253632354, updateTime=1776253632354, creator=18614031015, updator=18614031015), WebsiteProps(id=1251261682262163739, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1251257283494232110, code=grayFlag, value=0, createTime=1776253632369, updateTime=1776253632369, creator=18614031015, updator=18614031015), WebsiteProps(id=1251261682190860564, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1251257283494232110, code=logo, value=https://castjournals.cast.org.cn/joweb/wxdgc/CN/file/pic?fileId=AnvFcVcgnutwkVR+TFSYqg==, createTime=1776253632352, updateTime=1776253632352, creator=18614031015, updator=18614031015), WebsiteProps(id=1251261682283135261, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1251257283494232110, code=minRunFlag, value=0, createTime=1776253632374, updateTime=1776253632374, creator=18614031015, updator=18614031015), WebsiteProps(id=1251261682224414999, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1251257283494232110, code=picServerUrl, value=https://castjournals.cast.org.cn/joweb/wxdgc/CN/file/pic, createTime=1776253632360, updateTime=1776253632360, creator=18614031015, updator=18614031015), WebsiteProps(id=1251261682274746652, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1251257283494232110, code=silenceFlag, value=0, createTime=1776253632372, updateTime=1776253632372, creator=18614031015, updator=18614031015), WebsiteProps(id=1251261682207637782, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1251257283494232110, code=staticResourcePath, value=https://castjournals.cast.org.cn/joweb/cast_kjdb_cn_619/, createTime=1776253632356, updateTime=1776253632356, creator=18614031015, updator=18614031015), WebsiteProps(id=1251261682245386521, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1251257283494232110, code=themeColor, value=null, createTime=1776253632365, updateTime=1776253632365, creator=18614031015, updator=18614031015), WebsiteProps(id=1251261682249580826, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1251257283494232110, code=themeStyle, value=null, createTime=1776253632366, updateTime=1776253632366, creator=18614031015, updator=18614031015)]), Website(id=1251257283603284042, webName=null, webTitle=null, webDomain=null, webCopyrigh=null, webIpcNo=null, seoTitle=null, seoKeywords=null, seoDescription=null, tenantJournalId=null, journalId=1251234268282663017, 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/wxdgc/EN, language=EN, createTime=1776252583647, createBy=18614031015, updateTime=1776253597767, updateBy=18614031015, name=无线电工程-英文, tplId=1146101810881728533, title=Radio Engineering, delFlag=0, indexPage=/home, props=[WebsiteProps(id=1251261640977625682, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1251257283603284042, code=articleTextType, value=kx, createTime=1776253622526, updateTime=1776253622526, creator=18614031015, updator=18614031015), WebsiteProps(id=1251261640956654159, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1251257283603284042, code=banner, value=null, createTime=1776253622521, updateTime=1776253622521, creator=18614031015, updator=18614031015), WebsiteProps(id=1251261641002791509, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1251257283603284042, code=grayFlag, value=0, createTime=1776253622532, updateTime=1776253622532, creator=18614031015, updator=18614031015), WebsiteProps(id=1251261640948265550, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1251257283603284042, code=logo, value=https://castjournals.cast.org.cn/joweb/wxdgc/EN/file/pic?fileId=AnvFcVcgnutwkVR+TFSYqg==, createTime=1776253622519, updateTime=1776253622519, creator=18614031015, updator=18614031015), WebsiteProps(id=1251261641015374423, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1251257283603284042, code=minRunFlag, value=0, createTime=1776253622535, updateTime=1776253622535, creator=18614031015, updator=18614031015), WebsiteProps(id=1251261640969237073, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1251257283603284042, code=picServerUrl, value=https://castjournals.cast.org.cn/joweb/wxdgc/EN/file/pic, createTime=1776253622524, updateTime=1776253622524, creator=18614031015, updator=18614031015), WebsiteProps(id=1251261641011180118, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1251257283603284042, code=silenceFlag, value=0, createTime=1776253622534, updateTime=1776253622534, creator=18614031015, updator=18614031015), WebsiteProps(id=1251261640965042768, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1251257283603284042, code=staticResourcePath, value=https://castjournals.cast.org.cn/joweb/cast_kjdb_en_623/, createTime=1776253622523, updateTime=1776253622523, creator=18614031015, updator=18614031015), WebsiteProps(id=1251261640986014291, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1251257283603284042, code=themeColor, value=null, createTime=1776253622528, updateTime=1776253622528, creator=18614031015, updator=18614031015), WebsiteProps(id=1251261640994402900, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1251257283603284042, code=themeStyle, value=null, createTime=1776253622530, updateTime=1776253622530, creator=18614031015, updator=18614031015)])], journalTitle=无线电工程, weixinUrl=null, journalUrl=https://wxdg.cbpt.cnki.net/, iacademicId=null, status=1, seqNo=null, journalTitleEn=Radio Engineering, journalPhotoCn=g0Lcfmfo7oUYDLtQTsujLQ==, journalPhotoEn=fe/NRE6hx/Z5ZHweFTw2gA==, journalFirstLetter=R, 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), detailUrlCn=https://castjournals.cast.org.cn/joweb/wxdgc/CN/10.3969/j.issn.1003-3106.2025.11.006, detailUrlEn=https://castjournals.cast.org.cn/joweb/wxdgc/EN/10.3969/j.issn.1003-3106.2025.11.006, pdfUrlCn=https://castjournals.cast.org.cn/joweb/wxdgc/CN/PDF/10.3969/j.issn.1003-3106.2025.11.006, pdfUrlEn=https://castjournals.cast.org.cn/joweb/wxdgc/EN/PDF/10.3969/j.issn.1003-3106.2025.11.006, aliStartDate=null, aliEndDate=null, collectionFlag=false, citedCount=null, citedUrl=null, reference=null)
收藏切换
基于改进的YOLOv11甜菜田间杂草识别算法研究
收藏切换
PDF下载
周子健 , 刘强 *
无线电工程 | 信号与信息处理 2025,55(11): 2184-2194
收起
收藏切换
无线电工程 | 信号与信息处理 2025, 55(11): 2184-2194
基于改进的YOLOv11甜菜田间杂草识别算法研究
全屏
周子健, 刘强*
作者信息
  • 甘肃农业大学 信息科学技术学院,甘肃 兰州 730070
  • 周子健 男,(1998—),硕士研究生。主要研究方向:深度学习、图像识别。

通讯作者:

刘强 男,(1974—),硕士,教授,硕士生导师。主要研究方向:农业信息化。
Research on Algorithm for Weed Recognition in Beet Fields Based on Improved YOLOv11
Zijian ZHOU, Qiang LIU*
Affiliations
  • College of Information Science and Technology, Gansu Agricultural University, Lanzhou 730070, China
出版时间: 2025-11-05 doi: 10.3969/j.issn.1003-3106.2025.11.006
文章导航
收藏切换

针对复杂真实场景下甜菜田间杂草识别效率低、精确率低、小目标漏检等问题,提出一种基于YOLOv11模型改进的甜菜田间杂草识别算法。在主干网络引入PoolFormer模块和AKConv模块增强模型对全局语义信息的捕捉能力来提升检测精度,提高了在低分辨率图像和小物体中的检测效果。AKConv模块通过动态调整卷积核参数和形状,提升模型对生长形态不规则的甜菜与杂草的特征提取能力;PoolFormer模块可以很好地将相互遮盖的甜菜与杂草的边缘特征进行分割。在头部网络加入高级筛选-特征融合金字塔网络(High-level Screening Feature Pyramid Network,HS-FPN)模块增强多尺度融合效率,提升幼苗期甜菜与杂草的特征提取效率和速度。通过实验得出,改进YOLOv11模型的精确率、召回率、mAP@0.5、mAP@0.5:0.95相较于改进前分别提升了6.9%、7.8%、7.9%、7.8%。结果表明,在甜菜田间杂草识别上取得了显著的提升效果,为复杂场景中检测甜菜田间杂草提供了一个更可行的解决方案。

YOLOv11  /  甜菜  /  杂草识别  /  边缘特征分割  /  小目标检测  /  深度学习

An algorithm for weed recognition in beet fields based on improved YOLOv11 model is proposed to address the problems of low efficiency, low accuracy, and missed detection of small targets in complex real-world scenarios. The PoolFormer module and AKConv module are introduced into the backbone network to enhance the model's ability to capture global semantic information to improve detection accuracy, enhancing the detection performance in low resolution images and small objects. The AKConv module improves the feature extraction ability of the model for beets and weeds with irregular growth patterns by dynamically adjusting the convolution kernel parameters and shapes, while the PoolFormer module can effectively segment the edge features of beets and weeds that cover each other. Secondly, the High-level Screening Feature Pyramid Network (HS-FPN) module is added to the head network to enhance the efficiency of multi-scale fusion and improve the feature extraction efficiency and speed of beets and weeds during the seedling stage. Through experiments, it is found that the improved YOLOv11 model achieves increases of 6.9%, 7.8%, 7.9%, and 7.8% in precision, recall, mAP@0.5 and mAP@0.5: 0.95, respectively, compared to the original model. The results show that this algorithm has achieved significant improvement in weed recognition in beet fields, providing a more feasible solution for detecting weeds in beet fields in complex scenarios.

YOLOv11  /  beet  /  weed identification  /  edge feature segmentation  /  small target detection  /  deep learning
周子健, 刘强. 基于改进的YOLOv11甜菜田间杂草识别算法研究. 无线电工程, 2025 , 55 (11) : 2184 -2194 . DOI: 10.3969/j.issn.1003-3106.2025.11.006
Zijian ZHOU, Qiang LIU. Research on Algorithm for Weed Recognition in Beet Fields Based on Improved YOLOv11[J]. Radio Engineering, 2025 , 55 (11) : 2184 -2194 . DOI: 10.3969/j.issn.1003-3106.2025.11.006
随着国家对甜菜产业发展的重视,多年来不断扩大甜菜种植面积。然而在甜菜种植过程中,仍面临着一些问题,比如在甜菜幼苗期间杂草的生长会影响甜菜对营养、水分、光照的吸收和生长发育空间,导致产量下降并影响甜菜质量[1]。目前我国传统的田间杂草控制方法主要是人工、化学、机械手段。人工除草需要消耗大量的人力,并且效率较低;机械除草无法很好地应对生长分布不规则的杂草,并且很容易损害农作物;化学除草使用除草剂进行大面积喷洒可有效应对分布不同的杂草,但是除草剂类农药残留物不仅对农作物的生长造成破坏,还会影响土壤生态环境,使杂草产生抗药性,导致影响甜菜产量、质量等一系列问题[2]。因此,为了做到精准施药、精准除草,首先要解决的问题是精准检测识别杂草[3]。随着精准农业的发展,通过深度学习、图像处理进行田间杂草的精准检测和识别。而这些基于图像的深度学习算法能自动提取图像深层特征,完成作物与杂草的分类[4-7]
例如彭明霞等[8]通过融合特征网络金字塔(Feature Pyramid Network,FPN)的Faster R-CNN模型用于复杂环境下棉田杂草识别,提升了识别精度,为变量喷药等精准农业技术提供有效的参考;杨明轩等[9]在YOLOv5模型中融合了注意力机制与Swin Transformer Block解决了复杂背景下杂草被遮挡而导致漏检、错检等问题;罗柳茗等[10]通过对数据增强后利用YOLOv8模型对甘蔗田间杂草进行识别,提高了对杂草的识别精度;林宗缪等[11]通过在YOLOv8模型中加入了密集连接网络、使用WIoU损失函数等方法解决了在杂草识别检测过程中定位不准确、准确率不高的问题。
以上各个目标检测模型应用于各类作物田间杂草识别,为杂草识别的准确性和定位提供了有力支持。然而上述研究在检测精度上仍有上升空间,因此本研究基于Ultralytics团队最新的YOLOv11模型提出了一种改进的甜菜田间杂草识别算法,针对全生长阶段甜菜与杂草识别精度低以及相互遮盖的问题,提高甜菜田间杂草识别效率。
由于甜菜外观特征从叶丛形成期开始明显区别于杂草,而在甜菜幼苗期虽然与极个别杂草有相似之处,但可以通过其幼苗期的叶片形态、叶柄、生长姿态、茎基部颜色等特征来区别。因此为了更好地识别幼苗期的杂草与甜菜的差异,在检测头部分加入了HS-FPN模块。在主干网络用PoolFormer模块和AKConv模块将C3k2中的部分结构进行替换来提升模型对不规则特征的提取和分割堆叠物体边缘的能力。
本算法采用公开的数据集Lincolnbeet,分别包含在幼苗期和生长期的甜菜和杂草2个目标类共4400张图像。图像大小为1902 pixel×1080 pixel[12]。数据集包括txt文件后缀的适用于YOLO系列的Darknet数据集、json文件后缀的COCO数据集、xml文件后缀的VOC数据集。本研究将这些数据集图片随机分为训练集和验证集,形成可用于训练和验证测试的数据集。
YOLOv11是由Ultralytics团队于2024年9月发布的YOLO(You Only Look Once)系列新一代的目标检测算法。该模型延续了单阶段检测架构,整体结构分为Backbone(骨干网络)、Neck(特征融合层)和Head(检测头)三部分[13]。较前代YOLOv8有所升级的是:在主干网络和颈部网络中采用新的C3k2模块替代YOLOv8的C2f模块。与C2f模块相比,C3k2模块通常将输入的特征分为两部分,一部分通过标准卷积层直接进行传递,而另一部分则通过多个C3k结构或Bottleneck结构进行深度特征的提取(其中k为可调整卷积核大小,当超参数C3k为True时使用C3k结构,否则使用Bottleneck结构)。最终将两部分特征进行拼接,再通过一个1×1的标准卷积进行融合。该模块可以使用户自定义卷积核的大小,提升了模型的灵活性,保持了轻量化,提高了有效提取深层次特征的效率与能力;主干网络加入的C2PSA模块是在C2f模块的基础上增加了金字塔切片注意力机制(Pyramid Split Attention,PSA),这一机制是为了优化模型在多尺度特征处理中的有效性,增强特征提取能力,从而提升在遮挡和高密度场景下的检测能力;在头部网络中,将分类分支的2个常规卷积替换为深度可分离卷积DWConv,可使参数量和计算量较YOLOv8大幅度降低[14-17]。具体结构如图1所示。
本研究的算法在原始的YOLOv11的基础上做出3个改进,其基本结构如图2所示。原始YOLOv11的头部是解耦的,包括检测(回归框)头和分类头,并拆分成2个独立的卷积支路。本研究在检测支路的第一个CBS模块前加入了HS-FPN模块,通过多级特征融合和筛选机制优化模型对不同尺度目标的检测能力,提升小目标和复杂环境下的检测速度;在骨干网络部分,将C3k2中C3的子模块Bottleneck Bool模块替换为PoolFormer形成C3k2_PoolFormer,用池化实现高效特征混合,降低了模型计算的复杂度,提升了边缘分割效果;将C3k2模块中的Conv子模块全部替换为AKConv模块形成C3k2_AKConv。AKConv模块突破传统卷积核的固定采样形状和尺寸的限制,通过动态调整卷积核参数和形状,提升模型特征提取的能力。
标准卷积的采样位置是固定的,导致卷积只能提取当前窗口的局部信息,而且不能捕获其他位置的信息。为了扩大感受野,要通过堆叠多个卷积层或者池化操作,导致信息丢失与冗余计算[18-20]。可变形卷积(Deformable Convolution, DConv)算法通过卷积运算来学习偏移量,采样覆盖范围可扩展至原网格附近区域。该方法在一定程度上弥补了卷积运算的不足。但是,DConv在标准卷积的基础上还要额外学习偏移量参数,且DConv也是规则的采样网格,其卷积核大小与标准卷积相同,都是固定的K2,同样会产生很多的参数量[21]
因此本文算法使用AKConv,其结构如图3所示[22]。AKConv通过自适应全局动态调整采样形状,突破传统卷积的规则网格限制,实现更精准的特征提取,其线性的参数增长机制显著降低了计算开销。首先,输入维度为(C,H,W)(C表示通道数,HW表示图像的高度和宽度)的图像,并以左上角的采样点为采样原点;接着通过二维卷积来得到相应的偏移量,其维数为(B,2N,H,W)(B表示批处理大小,N表示卷积核的大小),同时将偏移量和原始坐标相加得到修改后的坐标;然后经过第一个Reshaple,用调整后的坐标通过双线性插值(Bilinear Interpolation,BI)获取不规则特征的采样位置,再通过卷积提取不规则采样位置对应的特征;最后,经过SiLU激活函数和归一化后得到最终的特征输出[23-24]
在这个过程中,由于调整后的采样位置可能呈现不规则分布,难以直接使用标准卷积提取特征。因此,为了有效提取不规则卷积核的采样位置对应的特征,可以使用多种方法,例如空间维度堆叠:将特征图在空间维度展开通过在行或列上堆叠特征,使用列卷积或行卷积提取不规则采样形状对应的特征,在提取特征时使用适当大小和步长的卷积核;通道维度堆叠:将N个采样点的特征沿通道维度堆叠为(C×N,H,W),并利用1×1卷积将维度降维到(C, H,W);3D卷积法:将维度展开为(C,N,H,W),然后使用步长和卷积大小为N×1×1的Conv3d来提取特征。以上提到的方法都可以提取出对应的不规则采样形状的特征,只需要对特征进行重塑,并使用相应的卷积运算即可[22]
HS-FPN是一种针对多尺度目标检测优化的特征金字塔网络架构,该网络由2个关键模块构成:特征选择模块和特征融合模块[25-27]
特征选择模块:通道注意力(Channel Attention,CA)和维度匹配(Dimension Matching,DM)在此模块中起着重要作用。其结构如图4所示,CA模块首先处理输入的特征图Fin∈R C×H×W,进行全局平均池化和全局最大池化,将得到的2个新的特征进行特征权相加后使用Sigmoid激活函数来确定每个通道的权重值,最终得到每个通道的权重。DM模块采用1×1卷积操作对输入的特征进行通道压缩处理,将其维度统一标准化为256通道,使不同通道数的特征进行有效融合,解决了多尺度特征图间的通道维度不匹配问题,保障跨尺度信息融合的有效性。
特征融合模块主要采用选择性特征融合(Selective Feature Fusion,SFF)机制来实现[28],结构如图5所示。SFF模块利用高层特征作为权重并结合CA模块对低层特征进行细化,有选择性地从低层特征中提取出重要的信息。给定输入高级特征Fhigh∈RC×H×W和输入低级特征,先将高层特征使用步长为2、卷积核大小为3×3的转置卷积(T-Conv)进行扩展,得到高级输入的新的维度Fhigh∈RC×2H×2W;然后,为了统一高级特征和低级特征的维度使用BI对上述高级特征进行上采样或下采样得到后通过CA模块将高级特征转化为相应的注意力权重,对低级特征进行过滤;最后将过滤后的低级特征与高级特征进行拼接,得到拼接后的特征图。SFF公式如下:
PoolFormer模块是基于MetaFormer模块这一范式架构的实例化模块,而MetaFormer通用架构是从TransFormer模块抽象出的核心架构。TransFormer架构包括输入、多个归一化模(Norm)、注意力模块(Attention)、通道多层感知机模块(Channel Multi-Layer Perceptron,Channel MLP)和残差链接。虽然TransFormer模块有全局信息建模、动态权重分配等强大性能,但是其注意力计算复杂度高,而且如果在小数据上训练容易发生过拟合等情况[29]。因此本研究采用PoolFormer架构,PoolFormer根据MetaFormer架构设计思想,将混合器令牌(token mixer)部分由Attention模块替换为了一个步长为1的池化块(Pooling)[30],其结构如图6所示。
PoolFormer结构:将数据输入后通过第一次归一化(Norm)来稳定特征分布,加快模型的收敛速度;将归一化后的数据通过池化模块进行平均池化;将得到的结果与原始数据进行第一次残差连接来有效缓解梯度消失问题,同时保留原始特征的低阶语义信息;进行第二次归一化处理;让数据通过MLP,该模块包含2个全连接层与非线性激活函数,并通过升维-降维策略来实现跨通道信息交互,此操作增强了模型非线性表达能力;将通过MLP模块的数据与第一次残差链接后的数据进行第二次残差链接,在保留初级特征的基础上逐步增强高阶语义表示。
池化块Pooling:Pooling模块的结构如图7所示。首先将输入的数据进行维度重排以符合Avg-Pool2d输入格式;其次对数据进行空间平均池化,来捕获邻近位置的特征关系,其池化窗口大小为3×3;然后将池化后的数据进行维度恢复;最后用池化后的数据减去原始输入来捕捉局部空间变化的特征,同时避免了单纯池化导致的信息损失。
这种改进虽然放弃了注意力机制,但由于使用无参池化操作从而降低了计算复杂度,而其动态权重分配功能则是通过第一次残差连接与归一化来进行的。同时体现出了在小数据集上训练的良好的鲁棒性[31]
本研究采用准确率(Precision,P)、召回率(Recall,R)、平均精度均值(mean Average Precision, mAP)作为模型的检测效果评价指标。P指模型正确识别出的目标占模型所有识别出的目标的比例。反映了模型的准确性,其值越高误检越少。R指成功识别的目标占实际应识别出的总目标的比例,作为衡量模型能识别出目标能力的度量,其值越高,漏检就越少,计算如下:
式中:TP表示模型识别的结果中正确的目标个数, FP表示模型识别的结果中错误识别的目标个数, FN表示模型未能识别的正确目标的个数。
mAP是衡量模型性能的综合指标,综合反映了模型在不同类别和检测难度下的整体表现。本研究采用mAP@0.5(IoU阈值为0.5)和mAP@0.5 :0.95(IoU阈值为0.5~0.95)两个指标来评估模型的综合指标。取值0~1,值越大表示模型性能越好,计算如下:
式中:N表示数据集中目标类别数。
本实验使用Python3.8开发环境、Pytorch深度学习网络框架。在训练过程中,超参数设置训练轮次为100、初始学习率为0.0005、批量大小为100。实验时间为2024年12月—2025年1月,实验地点为甘肃农业大学信息科学技术学院。在模型训练过程中使用自适应矩估计(Adaptive Moment Estimation,Adam)为优化器对训练过程进行优化。
为了验证模型3个改进点以目标检测有一定效果的提升,以原YOLOv11模型作为对比,分别对3个改进点进行了单一实验的有效性对比,并与改进后的YOLOv11模型也进行了对比。由表1可以看出,虽然原YOLOv11算法有较好的识别率,但是仍较逊于改进后的算法。
在主干网络应用AKConv模块和PoolFormer模块后,对杂草识别的精确率、召回率以及平均精度都有所提升。其中AKConv模块在精确率方面提升较高,较原模型提升了6.1%;PoolFormer模块在召回率方面提升较高,较原模型提升了5.6%。这说明AKConv模块通过动态调整卷积核参数和形状有效地提升了模型对甜菜与杂草的特征提取能力,提高了模型对杂草识别的精确率;而PoolFormer模块通过改进TransFormer将其中的Attention换成了Pooling,用多粒度结构分层捕获不同范围的上下文信息使相互遮盖的甜菜与杂草边缘可以被很好地分割,从而降低了模型对杂草的漏检率。
在头部网络加入HS-FPN模块,精确率提升了4.1%,召回率提升了2.9%,mAP@0.5和mAP@0.5:0.95都分别提升了3.9%,说明HS-FPN对处于幼苗期的甜菜与杂草的特征提取效率有所提升,因此提高了模型对幼苗期杂草的识别速度和精度。
最终改进后的模型相比原模型在各评价参数上都有提升,其中PoolFormer模块在mAP@ 0.5、mAP@0.5:0.95这2个指标上分别较原模型提升了6%和6.4%,略高于其他2个模块。由此可以看出,PoolFormer模块在改进的模型中贡献较大。而总体来看,改进后的模型其检测精度都有所提升,验证了这3个模块对甜菜田间杂草识别的可行性。
为了客观评估改进后的模型,对比了YOLOv9、YOLOv10、YOLOv11与改进模型在同一数据集上的性能指标,具体数据如表2所示。尽管YOLO模型在版本迭代的过程中其识别精度、效率等方面相较于旧模型得到了改进,但适配实际情况才是关键。因此通过对比包括YOLO算法的历史版本和本改进研究算法在内的4个算法,识别效率提升的原因是改进算法本身的贡献,而不是依靠底层技术的升级,也为改进算法在不同性能区间的提升幅度提出了依据。可以清晰地看出,本研究提出的算法在杂草目标检测中任务中具有明显的优势,改进的模型在精确率、召回率、mAP@0.5、mAP@0.5:0.95这4个指标上显著超越了YOLOv9和YOLOv10;与YOLOv11相比,改进的模型分别提升了6.9%、7.8%、7.9%、7.8%。
训练损失和验证损失下降过程如图8图9所示,可以看出,改进模型的边界框损失(box _loss)、分类损失(cls_loss)均低于其他对比模型。在前40个轮次模型损失快速减少并迅速拟合,在后60个轮次,模型损失趋于平稳并收敛,没有出现过拟合和欠拟合的情况。
本研究在原YOLOv11模型的基础进行改进,实现了动态调整卷积核形状,提升了不规则目标检测能力;利用多尺度特征筛选融合,优化了对小目标的检测效果。结合实验结果表明,改进的模型在检测效果上相对其他模型表现更好,正确识别和目标定位的能力更优。
为了能直观地体现本研究算法的可行性,对验证集中的图片进行验证,图10清晰地展示了本文算法通过在主干网络和头部网络的改进,图中0代表甜莱(beet),1代表杂草(weed),在幼苗期、相互遮挡的复杂背景下能够准确地检测出甜菜与杂草,在提高检测精度的同时降低了漏检率和误检率。
目前深度学习目标检测算法广泛应用于田间杂草识别,但现有模型对甜菜田间杂草识别任务存在识别精度低、小目标漏检等问题。因此本研究针对以上问题在真实场景下提出了一种基于YOLOv11改进的甜菜田间杂草识别算法。在主干网络引入PoolFormer模块和AKConv模块,以强化全局上下文建模,有效捕捉田间复杂背景下的特征提取并提高低分辨率下甜菜与杂草的纹理、形状、边缘等识别效率,提高了在相互遮挡情况下的识别精度。在检测头部分加入了HS-FPN模块用来降低幼苗期杂草的漏检率。通过使用公开数据集Lincolnbeet进行实验,实验结果表明改进后的YOLOv11模型在精确率、召回率、mAP@ 0.5、mAP@0.5:0.95分别达到了82%、77%、82%、60% ,相较于改进前的提升幅度分别为6.9%、7.8%、7.9%、8%。
改进后的YOLOv11算法在复杂环境下可以更准确地检测甜菜田间杂草,对精准农业和智慧农业发展具有一定的意义。但本研究在模型轻量化的优化方面还存在改进空间,未来的改进方向在于将模型轻量化以便在移动端部署,以期望模型在甜菜田间杂草识别任务中展现更好的应用价值和潜力。
参考文献 引证文献
排序方式:
[1]
沈涛,张秀再,许岱.改进RT-DETR的遥感图像小目标检测算法[J].计算机科学,2025,52(8):214-221.
[2]
胡炼,刘海龙,何杰,.智能除草机器人研究现状与展望[J].华南农业大学学报,2023,44(1):34-42.
[3]
袁洪波,赵努东,程曼.基于图像处理的田间杂草识别研究进展与展望[J].农业机械学报,2020,51(增刊2):323-334.
[4]
陈雪,谭峰.基于改进YOLOv5的轻量级大豆田间杂草识别方法研究[J].农机使用与维修,2025(2):1-7.
[5]
文韬,王天一,黄诗锐,.基于改进YOLOv8的农作物与藜草检测算法:MES-YOLO[J/OL].计算机工程与科学,1-13[2025-06-15]. https://link.cnki.net/urlid/43.1258.tp.20241011.1309.004.
[6]
杜希望,李霞,唐腾飞,.基于改进YOLOv8n的双模态杂草检测方法[J/OL].天津理工大学学报,1-10[2025-05-15]. https://link.cnki.net/urlid/12.1374.N.20241029.1920.024.
[7]
冀汶莉,刘洲,邢海花.基于YOLO v5的农田杂草识别轻量化方法研究[J].农业机械学报,2024,55(1):212-222.
[8]
彭明霞,夏俊芳,彭辉.融合FPN的Faster R-CNN复杂背景下棉田杂草高效识别方法[J].农业工程学报, 2019,35(20):202-209.
[9]
杨明轩,陈琳.改进YOLOv5的棉田杂草检测[J].现代电子技术,2024,47(24):60-67.
[10]
罗柳茗,李岩舟,石美琦,.基于YOLOv8n的甘蔗杂草识别方法[J].中国农机化学报,2025,46(2): 237-244.
[11]
林宗缪,马超,胡冬.基于改进YOLOv8卷积神经网络的稻田苗期杂草检测方法[J].湖北农业科学,2024, 63(8):17-22.
[12]
DARBYSHIRE M,SALAZAR-GOMEZ A,GAO J F,et al. Towards Practical Object Detection for Weed Spraying in Precision Agriculture[J]. Frontiers in Plant Science, 2023,14:1183277.
[13]
KHANAM R,HUSSAIN M. YOLOv11: An Overview of the Key Architectural Enhancements[EB/OL]. (2024-10-23)[2025-05-10]. https://arxiv.org/abs/2410.17725.
[14]
ZHANG R T,ZHU D J,WU C,et al. Sensitive Information Detection Based on Deep Learning Models[J]. Applied Sciences,2024,14(17):7541.
[15]
HIDAYATULLAH P,SYAKRANI N,SHOLAHUDDIN M R,et al. YOLOv8 to YOLO11: A Comprehensive Architecture In-depth Comparative Review[EB/OL]. (2025-01-23)[2025-05-11]. https://arxiv.org/abs/2501.13400.
[16]
马艺嘉,卢睿智,付可欣,.基于MBL-YOLOv8n模型的光伏缺陷检测[J].无线电工程,2025,55(8):1703-1709.
[17]
李璐,陈清江.基于超分辨率和YOLO的红外小目标检测[J].无线电工程,2025,55(8):1571-1579.
[18]
QIN J H,PAN W Y,XIANG X Y,et al. A Biological Image Classification Method Based on Improved CNN[J]. Ecological Informatics,2020,58:11093.
[19]
YANG E Q,ZHOU W J,QIAN X H,et al. DRNet: Dual-stage Refinement Network with Boundary Inference for RGB-D Semantic Segmentation of Indoor Scenes[J]. Engineering Applications of Artificial Intelligence, 2023, 125:106729.
[20]
WANG X, HE N, HONG C, et al. Improved YOLOX-X Based UAV Aerial Photography Object Detection Algorithm[J]. Image and Vision Computing,2023,135:104697.
[21]
ZHU X Z,HU H,LIN S,et al. Deformable ConvNets v2:More Deformable,Better Results[EB/OL]. (2018-11-27)[2025-05-11]. https://arxiv.org/abs/1811.11168.
[22]
ZHANG X,SONG Y Z,SONG T T,et al. AKConv: Convolutional Kernel with Arbitrary Sampled Shapes and Arbitrary Number of Parameters[EB/OL]. (2023-11-20)[2025-05-11]. https://arxiv.org/abs/2311.11587v1.
[23]
JIA Z Y, ZHANG M, YUAN C, et al. ADL-YOLOv8: A Field Crop Weed Detection Model Based on Improved YOLOv8[J]. Agronomy,2024,14(10):2355.
[24]
SUN H,TAN C H,PANG S,et al. RA-YOLOv8: An Improved YOLOv8 Seal Text Detection Method[J]. Electronics,2024,13(15):3001.
[25]
CHEN Y F,ZHANG C Y,CHEN B,et al. Accurate Leukocyte Detection Based on Deformable-DETR and Multilevel Feature Fusion for Aiding Diagnosis of Blood Diseases[EB/OL]. (2024-01-01)[2025-05-11]. https://arxiv.org/abs/2401.00926.
[26]
LIU R M,SU W H. APHS-YOLO: A Lightweight Model for Real-time Detection and Classification of Stropharia Rugoso-Annulata[J]. Foods,2024,13(11):1710.
[27]
XIN D Y,LI T Q. Revolutionizing Tomato Disease Detection in Complex Environments[J]. Frontiers in Plant Science,2024,15:1409544.
[28]
MA X W,YANG J W,HONG T F,et al. STNet: Spatial and Temporal Feature Fusion Network For Change Detection in Remote Sensing Images[C]//2023 IEEE International Conference on Multimedia and Expo (ICME). Brisbane:IEEE,2023:2195-2200.
[29]
VASWANI A,SHAZEER N,PARMAR N,et al. Attention Is All You Need[EB/OL]. (2017-06-12)[2025-05-12]. https://arxiv.org/abs/1706.03762.
[30]
YU W H,LUO M,ZHOU P,et al. MetaFormer Is Actually What You Need for Vision[EB/OL]. (2021-11-12)[2025-05-10]. https://arxiv.org/abs/2111.11418.
[31]
LIAO Y N,HUANG C Y,ZHANG T,et al. China 05 Bone Age Assessment Based on Deep Learning[J]. The Journal of China Universities of Posts and Telecommunications, 2024,31(6):66-75.
2025年第55卷第11期
PDF下载
137
66
引用本文
BibTeX
文章信息
doi: 10.3969/j.issn.1003-3106.2025.11.006
  • 接收时间:2025-07-01
  • 首发时间:2026-04-17
  • 出版时间:2025-11-05
补充材料
相关文章
文章信息
作者
出版历史
  • 收稿日期:2025-07-01
基金
作者信息
    甘肃农业大学 信息科学技术学院,甘肃 兰州 730070

通讯作者:

刘强 男,(1974—),硕士,教授,硕士生导师。主要研究方向:农业信息化。
参考文献
分享链接
https://castjournals.cast.org.cn/joweb/wxdgc/CN/10.3969/j.issn.1003-3106.2025.11.006
分享至
全文二维码

扫描看全文

引用本文
BibTeX
本文的引用情况
2种不同金属材料的力学参数

Family
属数
Number of
genus
种数
Number of
species
占总种数比例
Percentage of
total species (%)

Genus
种数
Number of
species
占总种数比例
Percentage of total
species (%)
鹅膏菌科Amanitaceae 2 11 5.26 鹅膏菌属 Amanita 10 4.78
小菇科 Mycenaceae 2 12 5.74 丝盖伞属 Inocybe 5 2.39
多孔菌科 Polyporaceae 8 14 6.70 蜡蘑属 Laccaria 5 2.39
红菇科 Russulaceae 3 23 11.00 小皮伞属 Marasmius 6 2.87
小菇属 Mycena 11 5.26
光柄菇属 Pluteus 5 2.39
红菇属 Russula 17 8.13
栓菌属 Trametes 5 2.39
关闭全屏