Article(id=1312006310527332698, tenantId=1146029695717560320, journalId=1311385739772133495, issueId=1311999647825101743, articleNumber=null, orderNo=null, doi=10.6041/j.issn.1000-1298.2025.12.054, pmid=null, cstr=null, oa=null, hot=null, price=null, onlineType=0, articleFormat=0, articleType=null, articleTypeStr=null, receivedDate=1723392000000, receivedDateStr=2024-08-12, revisedDate=1729180800000, revisedDateStr=2024-10-18, acceptedDate=null, acceptedDateStr=null, onlineDate=1790736280328, onlineDateStr=2026-09-30, pubDate=1766592000000, pubDateStr=2025-12-25, doiRegisterDate=null, doiRegisterDateStr=null, onlineIssueDate=1790736280328, onlineIssueDateStr=2026-09-30, onlineJustAcceptDate=null, onlineJustAcceptDateStr=null, onlineFirstDate=null, onlineFirstDateStr=null, sourceXml=null, magXml=null, createTime=1790736280328, creator=13701087609, updateTime=1790736280328, updator=13701087609, issue=Issue{id=1311999647825101743, tenantId=1146029695717560320, journalId=1311385739772133495, year='2025', volume='56', issue='12', pageStart='1', pageEnd='818', issueExtLink='null', onlineDate='null', pubDate='1766592000000', pubDateStr='2025-12-25', beforeIssueId=null, nextIssueId=null, price=null, status=1, issueComplete=1, articleOrder=1, issueType=1, specialIssue=null, createTime=1790734691815, creator='13701087609', updateTime=1790738411040, updator='13701087609', preIssue=null, nextIssue=null, articleTotal=null, ext={EN=IssueExt(id=1312015247465083866, tenantId=1146029695717560320, journalId=1311385739772133495, issueId=1311999647825101743, language=EN, specialIssueTitle=, coverIllustrator=null, specialIssueEditor=, specialIssueAbout=), CN=IssueExt(id=1312015247465083867, tenantId=1146029695717560320, journalId=1311385739772133495, issueId=1311999647825101743, language=CN, specialIssueTitle=, coverIllustrator=null, specialIssueEditor=, specialIssueAbout=)}, issueFiles=null, downloadFileDto=null}, startPage=591, endPage=602, ext={EN=ArticleExt(id=1312006310741242203, articleId=1312006310527332698, tenantId=1146029695717560320, journalId=1311385739772133495, language=EN, title=Individual Identification Method of Simmental Cattle Based on Improved YOLO v8 Keypoint Detection, columnId=1311999800569066336, journalTitle=Transactions of the Chinese Society for Agricultural Machinery, columnName=Agricultural Informatization Engineering, runingTitle=null, highlight=null, articleAbstract=

With the intelligent development of animal husbandry, computer-vision-based individual cattle recognition is playing an increasingly important role in modern cattle farming. However, in practical applications, the rapid changes in the angles of cattle faces, frequent fluctuations in the number of cattle, and limitations in model recognition speed severely affect the accuracy and efficiency of recognition. To address these challenges, an individual recognition method for Simmental cattle was proposed based on improved YOLO v8 keypoint detection. The method consisted of two stages: in the cattle face detection stage, the improved YOLO v8 model automatically detected keypoints in cattle face images and performed correction alignment and background cropping to make the angles of the cattle face images more consistent. In the cattle face recognition stage, depthwise separable convolution was firstly introduced to reduce the number of parameters, and then residual connections and attention mechanisms were utilized to enhance the model's feature extraction capabilities, thereby improving recognition accuracy and speed. To validate the effectiveness of the method, a dataset of cattle face images containing 159 Simmental cattle was constructed. Experimental results showed that the proposed model had significant improvements in the number of parameters and inference speed, with processing time for a single image reduced to within 39 ms and a recognition accuracy of 95.8%, an average improvement of 4.8 percentage points compared with that of other models. Additionally, the proposed method supported real-time updates to the cattle database, maintaining high accuracy even after adding new cattle face images.

, authors=Zhenjiang TIAN1, Cheng HAN1, Chao ZHANG1, Yueying LI1, Fudong YU2, Xinhui ZHONG2, authorsList=Zhenjiang TIAN, Cheng HAN, Chao ZHANG, Yueying LI, Fudong YU, Xinhui ZHONG, authorCompany=null, correspAuthors=Cheng HAN, 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, fund=null), CN=ArticleExt(id=1312006314163794306, articleId=1312006310527332698, tenantId=1146029695717560320, journalId=1311385739772133495, language=CN, title=基于改进YOLO v8关键点检测的西门塔尔牛只个体识别方法, columnId=1311999800749421410, journalTitle=农业机械学报, columnName=农业信息化工程, runingTitle=null, highlight=null, articleAbstract=

随着畜牧业的智能化发展,基于计算机视觉技术的牛只个体识别在牛场养殖中扮演着越来越重要的角色。然而,在实际应用中,牛脸角度的快速变化、牛只数量的频繁变动以及模型识别速度的局限性严重影响了识别的准确性和效率。为了应对这些挑战,本文提出了一种基于改进YOLO v8关键点检测的西门塔尔牛个体识别方法。该方法包括2个阶段,在牛脸检测阶段,改进YOLO v8模型自动检测牛脸图像中的关键点,并进行矫正对齐和背景裁剪,使牛脸图像角度趋于一致。在牛脸识别阶段,引入了深度可分离卷积以减少参数量,利用残差连接和注意力机制增强了模型的特征提取能力,从而提升了识别准确性和速度。为验证方法的有效性,构建了包含159头西门塔尔牛的牛脸图像数据集。实验结果表明,本文模型在参数量和推理时间上有显著改进,单幅图像处理时间缩短至39 ms以内,识别准确率为95.8%,比其他模型相比平均提高4.8个百分点。此外,本文方法支持实时更新牛只数据库,在增加牛脸图像后,仍能保持较高准确率。

, authors=田振江1, 韩成1, 张超1, 李月莹1, 于福东2, 钟馨慧2, authorsList=田振江, 韩成, 张超, 李月莹, 于福东, 钟馨慧, authorCompany=null, correspAuthors=韩成, authorNote=

田振江(1998—),男,博士生,主要从事牛脸面部识别研究,E-mail:

, correspAuthorsNote=
韩成(1978—),男,教授,主要从事计算机视觉研究,E-mail:
, copyrightStatement=null, copyrightOwner=null, extLink=null, articleAbsUrl=null, sourceXml=9V8FrugwNcsXywUuI/ti4g==, magXml=NdRrUA+iUxVYSFK55OHLsw==, pdfUrl=null, pdf=BSlagxBUAx+51H71loaINA==, pdfFileSize=11388615, pdfExtLink=null, richHtmlUrl=null, mobilePdfUrl=null, reviewReport=null, pdfFirstPage=null, abstractGraph=1ztwDc9//k58LPtrjgIl8w==, abstractGraphContent=null, abstractVideo=null, citation=null, cebUrl=null, magXmlContent=D4ATvSho3BIAFb8rnrwoMQ==, mapNumber=null, fund=null)}, authors=[Author(id=1312006314591613322, tenantId=1146029695717560320, journalId=1311385739772133495, articleId=1312006310527332698, orderNo=0, firstName=null, middleName=null, lastName=null, nameCn=null, orcid=null, stid=null, country=null, authorPic=null, dead=0, email=2022101001@mails.cust.edu.cn, emailSecond=null, emailThird=null, correspondingAuthor=0, authorType=1, ext={EN=AuthorExt(id=1312006314679693708, tenantId=1146029695717560320, journalId=1311385739772133495, articleId=1312006310527332698, authorId=1312006314591613322, language=EN, stringName=Zhenjiang TIAN, firstName=Zhenjiang, middleName=null, lastName=TIAN, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, address=1.School of Computer Science and Technology, Changchun University of Science and Technology, Changchun 130012, China, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null), CN=AuthorExt(id=1312006314776162701, tenantId=1146029695717560320, journalId=1311385739772133495, articleId=1312006310527332698, authorId=1312006314591613322, language=CN, stringName=田振江, firstName=null, middleName=null, lastName=null, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, address=1.长春理工大学计算机科学技术学院,长春 130012, bio={"content":"

田振江(1998—),男,博士生,主要从事牛脸面部识别研究,E-mail:

"}, bioImg=null, bioContent=

田振江(1998—),男,博士生,主要从事牛脸面部识别研究,E-mail:

, aboutCorrespAuthor=null)}, companyList=[AuthorCompany(id=1312006314411258243, tenantId=1146029695717560320, journalId=1311385739772133495, articleId=1312006310527332698, xref=1., ext=[AuthorCompanyExt(id=1312006314419646852, tenantId=1146029695717560320, journalId=1311385739772133495, articleId=1312006310527332698, companyId=1312006314411258243, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1.School of Computer Science and Technology, Changchun University of Science and Technology, Changchun 130012, China), AuthorCompanyExt(id=1312006314423841157, tenantId=1146029695717560320, journalId=1311385739772133495, articleId=1312006310527332698, companyId=1312006314411258243, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1.长春理工大学计算机科学技术学院,长春 130012)])]), Author(id=1312006314847465871, tenantId=1146029695717560320, journalId=1311385739772133495, articleId=1312006310527332698, orderNo=1, firstName=null, middleName=null, lastName=null, nameCn=null, orcid=null, stid=null, country=null, authorPic=null, dead=0, email=hancheng@cust.edu.cn, emailSecond=null, emailThird=null, correspondingAuthor=1, authorType=1, ext={EN=AuthorExt(id=1312006314927157649, tenantId=1146029695717560320, journalId=1311385739772133495, articleId=1312006310527332698, authorId=1312006314847465871, language=EN, stringName=Cheng HAN, firstName=Cheng, middleName=null, lastName=HAN, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, address=1.School of Computer Science and Technology, Changchun University of Science and Technology, Changchun 130012, China, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null), CN=AuthorExt(id=1312006314998460818, tenantId=1146029695717560320, journalId=1311385739772133495, articleId=1312006310527332698, authorId=1312006314847465871, language=CN, stringName=韩成, firstName=null, middleName=null, lastName=null, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, address=1.长春理工大学计算机科学技术学院,长春 130012, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null)}, companyList=[AuthorCompany(id=1312006314411258243, tenantId=1146029695717560320, journalId=1311385739772133495, articleId=1312006310527332698, xref=1., ext=[AuthorCompanyExt(id=1312006314419646852, tenantId=1146029695717560320, journalId=1311385739772133495, articleId=1312006310527332698, companyId=1312006314411258243, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1.School of Computer Science and Technology, Changchun University of Science and Technology, Changchun 130012, China), AuthorCompanyExt(id=1312006314423841157, tenantId=1146029695717560320, journalId=1311385739772133495, articleId=1312006310527332698, companyId=1312006314411258243, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1.长春理工大学计算机科学技术学院,长春 130012)])]), Author(id=1312006315061375380, tenantId=1146029695717560320, journalId=1311385739772133495, articleId=1312006310527332698, 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=1312006315132678550, tenantId=1146029695717560320, journalId=1311385739772133495, articleId=1312006310527332698, authorId=1312006315061375380, language=EN, stringName=Chao ZHANG, firstName=Chao, middleName=null, lastName=ZHANG, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, address=1.School of Computer Science and Technology, Changchun University of Science and Technology, Changchun 130012, China, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null), CN=AuthorExt(id=1312006315220758935, tenantId=1146029695717560320, journalId=1311385739772133495, articleId=1312006310527332698, authorId=1312006315061375380, language=CN, stringName=张超, firstName=null, middleName=null, lastName=null, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, address=1.长春理工大学计算机科学技术学院,长春 130012, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null)}, companyList=[AuthorCompany(id=1312006314411258243, tenantId=1146029695717560320, journalId=1311385739772133495, articleId=1312006310527332698, xref=1., ext=[AuthorCompanyExt(id=1312006314419646852, tenantId=1146029695717560320, journalId=1311385739772133495, articleId=1312006310527332698, companyId=1312006314411258243, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1.School of Computer Science and Technology, Changchun University of Science and Technology, Changchun 130012, China), AuthorCompanyExt(id=1312006314423841157, tenantId=1146029695717560320, journalId=1311385739772133495, articleId=1312006310527332698, companyId=1312006314411258243, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1.长春理工大学计算机科学技术学院,长春 130012)])]), Author(id=1312006315283673497, tenantId=1146029695717560320, journalId=1311385739772133495, articleId=1312006310527332698, 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=1312006315367559579, tenantId=1146029695717560320, journalId=1311385739772133495, articleId=1312006310527332698, authorId=1312006315283673497, language=EN, stringName=Yueying LI, firstName=Yueying, middleName=null, lastName=LI, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, address=1.School of Computer Science and Technology, Changchun University of Science and Technology, Changchun 130012, China, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null), CN=AuthorExt(id=1312006315438862748, tenantId=1146029695717560320, journalId=1311385739772133495, articleId=1312006310527332698, authorId=1312006315283673497, language=CN, stringName=李月莹, firstName=null, middleName=null, lastName=null, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, address=1.长春理工大学计算机科学技术学院,长春 130012, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null)}, companyList=[AuthorCompany(id=1312006314411258243, tenantId=1146029695717560320, journalId=1311385739772133495, articleId=1312006310527332698, xref=1., ext=[AuthorCompanyExt(id=1312006314419646852, tenantId=1146029695717560320, journalId=1311385739772133495, articleId=1312006310527332698, companyId=1312006314411258243, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1.School of Computer Science and Technology, Changchun University of Science and Technology, Changchun 130012, China), AuthorCompanyExt(id=1312006314423841157, tenantId=1146029695717560320, journalId=1311385739772133495, articleId=1312006310527332698, companyId=1312006314411258243, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1.长春理工大学计算机科学技术学院,长春 130012)])]), Author(id=1312006315505971614, tenantId=1146029695717560320, journalId=1311385739772133495, articleId=1312006310527332698, 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=1312006315577274784, tenantId=1146029695717560320, journalId=1311385739772133495, articleId=1312006310527332698, authorId=1312006315505971614, language=EN, stringName=Fudong YU, firstName=Fudong, middleName=null, lastName=YU, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=2, address=2.Jilin Sino Agri Sunshine Data Co., Ltd., Changchun 130022, China, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null), CN=AuthorExt(id=1312006315648577953, tenantId=1146029695717560320, journalId=1311385739772133495, articleId=1312006310527332698, authorId=1312006315505971614, language=CN, stringName=于福东, firstName=null, middleName=null, lastName=null, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=2, address=2.吉林省中农阳光数据有限公司,长春 130022, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null)}, companyList=[AuthorCompany(id=1312006314499338630, tenantId=1146029695717560320, journalId=1311385739772133495, articleId=1312006310527332698, xref=2., ext=[AuthorCompanyExt(id=1312006314503532935, tenantId=1146029695717560320, journalId=1311385739772133495, articleId=1312006310527332698, companyId=1312006314499338630, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=2.Jilin Sino Agri Sunshine Data Co., Ltd., Changchun 130022, China), AuthorCompanyExt(id=1312006314511921544, tenantId=1146029695717560320, journalId=1311385739772133495, articleId=1312006310527332698, companyId=1312006314499338630, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=2.吉林省中农阳光数据有限公司,长春 130022)])]), Author(id=1312006315719881123, tenantId=1146029695717560320, journalId=1311385739772133495, articleId=1312006310527332698, 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=1312006315795378597, tenantId=1146029695717560320, journalId=1311385739772133495, articleId=1312006310527332698, authorId=1312006315719881123, language=EN, stringName=Xinhui ZHONG, firstName=Xinhui, middleName=null, lastName=ZHONG, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=2, address=2.Jilin Sino Agri Sunshine Data Co., Ltd., Changchun 130022, China, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null), CN=AuthorExt(id=1312006315891847590, tenantId=1146029695717560320, journalId=1311385739772133495, articleId=1312006310527332698, authorId=1312006315719881123, language=CN, stringName=钟馨慧, firstName=null, middleName=null, lastName=null, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=2, address=2.吉林省中农阳光数据有限公司,长春 130022, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null)}, companyList=[AuthorCompany(id=1312006314499338630, tenantId=1146029695717560320, journalId=1311385739772133495, articleId=1312006310527332698, xref=2., ext=[AuthorCompanyExt(id=1312006314503532935, tenantId=1146029695717560320, journalId=1311385739772133495, articleId=1312006310527332698, companyId=1312006314499338630, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=2.Jilin Sino Agri Sunshine Data Co., Ltd., Changchun 130022, China), AuthorCompanyExt(id=1312006314511921544, tenantId=1146029695717560320, journalId=1311385739772133495, articleId=1312006310527332698, companyId=1312006314499338630, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=2.吉林省中农阳光数据有限公司,长春 130022)])])], keywords=[Keyword(id=1312006315984122279, tenantId=1146029695717560320, journalId=1311385739772133495, articleId=1312006310527332698, language=EN, orderNo=1, keyword=cow face recognition), Keyword(id=1312006316055425448, tenantId=1146029695717560320, journalId=1311385739772133495, articleId=1312006310527332698, language=EN, orderNo=2, keyword=computer vision), Keyword(id=1312006316122534313, tenantId=1146029695717560320, journalId=1311385739772133495, articleId=1312006310527332698, language=EN, orderNo=3, keyword=YOLO v8), Keyword(id=1312006316198031786, tenantId=1146029695717560320, journalId=1311385739772133495, articleId=1312006310527332698, language=EN, orderNo=4, keyword=key point detection), Keyword(id=1312006316290306475, tenantId=1146029695717560320, journalId=1311385739772133495, articleId=1312006310527332698, language=CN, orderNo=1, keyword=牛脸识别), Keyword(id=1312006316374192556, tenantId=1146029695717560320, journalId=1311385739772133495, articleId=1312006310527332698, language=CN, orderNo=2, keyword=计算机视觉), Keyword(id=1312006316462272941, tenantId=1146029695717560320, journalId=1311385739772133495, articleId=1312006310527332698, language=CN, orderNo=3, keyword=YOLO v8), Keyword(id=1312006316537770414, tenantId=1146029695717560320, journalId=1311385739772133495, articleId=1312006310527332698, language=CN, orderNo=4, keyword=关键点检测)], refs=[Reference(id=1312006319385702864, tenantId=1146029695717560320, journalId=1311385739772133495, articleId=1312006310527332698, doi=null, pmid=null, pmcid=null, year=2024, volume=55, issue=3, pageStart=282, pageEnd=289, 391, url=null, language=null, rfNumber=[1], rfOrder=0, authorNames=张姝瑾, 许兴时, 邓洪兴, journalName=农业机械学报, refType=null, unstructuredReference=张姝瑾, 许兴时, 邓洪兴, 等. 基于YOLO v8n-seg-FCA-BiFPN的奶牛身体分割方法[J]. 农业机械学报, 2024, 55(3): 282-289, 391., articleTitle=基于YOLO v8n-seg-FCA-BiFPN的奶牛身体分割方法, refAbstract=null), Reference(id=1312006319494754769, tenantId=1146029695717560320, journalId=1311385739772133495, articleId=1312006310527332698, doi=null, pmid=null, pmcid=null, year=2024, volume=55, issue=3, pageStart=282, pageEnd=289, 391, url=null, language=null, rfNumber=[1], rfOrder=1, authorNames=ZHANG Shujin, XU Xingshi, DENG Hongxing, journalName=Transactions of the Chinese Society for Agricultural Machinery, refType=null, unstructuredReference=ZHANG Shujin, XU Xingshi, DENG Hongxing, et al. Segmentation model of cow body parts based on YOLO v8n-seg-FCA-BiFPN[J]. Transactions of the Chinese Society for Agricultural Machinery, 2024, 55(3): 282-289, 391. (in Chinese), articleTitle=Segmentation model of cow body parts based on YOLO v8n-seg-FCA-BiFPN, refAbstract=null), Reference(id=1312006319578640850, tenantId=1146029695717560320, journalId=1311385739772133495, articleId=1312006310527332698, doi=null, pmid=null, pmcid=null, year=2023, volume=30, issue=4, pageStart=68, pageEnd=76, url=null, language=null, rfNumber=[2], rfOrder=2, authorNames=GUO Y, HONG W, WU J, journalName=IEEE Robotics & Automation Magazine, refType=null, unstructuredReference=GUO Y, HONG W, WU J, et al. Vision-based cow tracking and feeding monitoring for autonomous livestock farming: the YOLOv5s-CA+DeepSORT-Vision Transformer[J]. IEEE Robotics & Automation Magazine, 2023, 30(4): 68-76., articleTitle=Vision-based cow tracking and feeding monitoring for autonomous livestock farming: the YOLOv5s-CA+DeepSORT-Vision Transformer, refAbstract=null), Reference(id=1312006319658332627, tenantId=1146029695717560320, journalId=1311385739772133495, articleId=1312006310527332698, doi=null, pmid=null, pmcid=null, year=2024, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=[3], rfOrder=3, authorNames=董正奇, journalName=null, refType=null, unstructuredReference=董正奇. 基于体温监测和饮水行为记录的牛只监管耳标设计[D]. 杨凌: 西北农林科技大学, 2024., articleTitle=基于体温监测和饮水行为记录的牛只监管耳标设计, refAbstract=null), Reference(id=1312006319725441492, tenantId=1146029695717560320, journalId=1311385739772133495, articleId=1312006310527332698, doi=null, pmid=null, pmcid=null, year=2024, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=[3], rfOrder=4, authorNames=DONG Zhengqi, journalName=null, refType=null, unstructuredReference=DONG Zhengqi. Cow regulatory ear tag design based on body temperature monitoring and drinking behavior records[D]. Yangling: Northwest A&F University, 2024. (in Chinese), articleTitle=Cow regulatory ear tag design based on body temperature monitoring and drinking behavior records, refAbstract=null), Reference(id=1312006319788356053, tenantId=1146029695717560320, journalId=1311385739772133495, articleId=1312006310527332698, doi=null, pmid=null, pmcid=null, year=2023, volume=54, issue=6, pageStart=272, pageEnd=279, url=null, language=null, rfNumber=[4], rfOrder=5, authorNames=司永胜, 王朝阳, 张艳, journalName=农业机械学报, refType=null, unstructuredReference=司永胜, 王朝阳, 张艳, 等. 基于深度与传统特征融合的非限制条件下奶牛个体识别[J]. 农业机械学报, 2023, 54(6): 272-279., articleTitle=基于深度与传统特征融合的非限制条件下奶牛个体识别, refAbstract=null), Reference(id=1312006319855464918, tenantId=1146029695717560320, journalId=1311385739772133495, articleId=1312006310527332698, doi=null, pmid=null, pmcid=null, year=2023, volume=54, issue=6, pageStart=272, pageEnd=279, url=null, language=null, rfNumber=[4], rfOrder=6, authorNames=SI Yongsheng, WANG Zhaoyang, ZHANG Yan, journalName=Transactions of the Chinese Society for Agricultural Machinery, refType=null, unstructuredReference=SI Yongsheng, WANG Zhaoyang, ZHANG Yan, et al. Individual identification of dairy cows under unrestricted conditions based on fusion of deep and traditional features[J]. Transactions of the Chinese Society for Agricultural Machinery, 2023, 54(6): 272-279. (in Chinese), articleTitle=Individual identification of dairy cows under unrestricted conditions based on fusion of deep and traditional features, refAbstract=null), Reference(id=1312006319943545303, tenantId=1146029695717560320, journalId=1311385739772133495, articleId=1312006310527332698, doi=null, pmid=null, pmcid=null, year=1943, volume=26, issue=2, pageStart=60, pageEnd=69, url=null, language=null, rfNumber=[5], rfOrder=7, authorNames=LOPEZ R B, journalName=Clinical and Experimental Optometry, refType=null, unstructuredReference=LOPEZ R B. Diseases of the heart and blood vessels and their relation to the eye[J]. Clinical and Experimental Optometry, 1943, 26(2): 60-69., articleTitle=Diseases of the heart and blood vessels and their relation to the eye, refAbstract=null), Reference(id=1312006320010654168, tenantId=1146029695717560320, journalId=1311385739772133495, articleId=1312006310527332698, doi=null, pmid=null, pmcid=null, year=2018, volume=46, issue=5, pageStart=812, pageEnd=837, url=null, language=null, rfNumber=[6], rfOrder=8, authorNames=KUMAR S, SINGH S K, ABIDI A I, journalName=International Journal of Parallel Programming, refType=null, unstructuredReference=KUMAR S, SINGH S K, ABIDI A I, et al. Group sparse representation approach for recognition of cattle on muzzle point images[J]. International Journal of Parallel Programming, 2018, 46(5): 812-837., articleTitle=Group sparse representation approach for recognition of cattle on muzzle point images, refAbstract=null), Reference(id=1312006320090345945, tenantId=1146029695717560320, journalId=1311385739772133495, articleId=1312006310527332698, doi=null, pmid=null, pmcid=null, year=2008, volume=116, issue=1-3, pageStart=42, pageEnd=52, url=null, language=null, rfNumber=[7], rfOrder=9, authorNames=ALLEN A, GOLDEN B, TAYLOR M, journalName=Livestock Science, refType=null, unstructuredReference=ALLEN A, GOLDEN B, TAYLOR M, et al. Evaluation of retinal imaging technology for the biometric identification of bovine animals in Northern Ireland[J]. Livestock Science, 2008, 116(1-3): 42-52., articleTitle=Evaluation of retinal imaging technology for the biometric identification of bovine animals in Northern Ireland, refAbstract=null), Reference(id=1312006320157454810, tenantId=1146029695717560320, journalId=1311385739772133495, articleId=1312006310527332698, doi=null, pmid=null, pmcid=null, year=2009, volume=19, issue=3, pageStart=155, pageEnd=160, url=null, language=null, rfNumber=[8], rfOrder=10, authorNames=孔强, 赵林度, journalName=中国安全科学学报, refType=null, unstructuredReference=孔强, 赵林度. 虹膜识别在肉类食品安全追溯系统中的应用及关键技术研究[J]. 中国安全科学学报, 2009, 19(3): 155-160., articleTitle=虹膜识别在肉类食品安全追溯系统中的应用及关键技术研究, refAbstract=null), Reference(id=1312006320228757979, tenantId=1146029695717560320, journalId=1311385739772133495, articleId=1312006310527332698, doi=null, pmid=null, pmcid=null, year=2009, volume=19, issue=3, pageStart=155, pageEnd=160, url=null, language=null, rfNumber=[8], rfOrder=11, authorNames=KONG Qiang, ZHAO Lindu, journalName=China Safety Science Journal, refType=null, unstructuredReference=KONG Qiang, ZHAO Lindu. Research on the application of iris recognition in meat-food supply chain management[J]. China Safety Science Journal, 2009, 19(3): 155-160. (in Chinese), articleTitle=Research on the application of iris recognition in meat-food supply chain management, refAbstract=null), Reference(id=1312006320291672540, tenantId=1146029695717560320, journalId=1311385739772133495, articleId=1312006310527332698, doi=null, pmid=null, pmcid=null, year=2010, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=[9], rfOrder=12, authorNames=盛大玮, journalName=null, refType=null, unstructuredReference=盛大玮. 牛眼虹膜识别技术研究[D]. 上海: 华东师范大学, 2010., articleTitle=牛眼虹膜识别技术研究, refAbstract=null), Reference(id=1312006320383947229, tenantId=1146029695717560320, journalId=1311385739772133495, articleId=1312006310527332698, doi=null, pmid=null, pmcid=null, year=2010, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=[9], rfOrder=13, authorNames=SHENG Dawei, journalName=null, refType=null, unstructuredReference=SHENG Dawei. Research on technology of cattle's iris recognition[D]. Shanghai: East China Normal University, 2010. (in Chinese), articleTitle=Research on technology of cattle's iris recognition, refAbstract=null), Reference(id=1312006320455250398, tenantId=1146029695717560320, journalId=1311385739772133495, articleId=1312006310527332698, doi=null, pmid=null, pmcid=null, year=2011, volume=21, issue=3, pageStart=124, pageEnd=130, url=null, language=null, rfNumber=[10], rfOrder=14, authorNames=李超, 赵林度, journalName=中国安全科学学报, refType=null, unstructuredReference=李超, 赵林度. 牛眼虹膜定位算法研究及其在肉食品追溯系统中的应用[J]. 中国安全科学学报, 2011, 21(3): 124-130., articleTitle=牛眼虹膜定位算法研究及其在肉食品追溯系统中的应用, refAbstract=null), Reference(id=1312006320551719391, tenantId=1146029695717560320, journalId=1311385739772133495, articleId=1312006310527332698, doi=null, pmid=null, pmcid=null, year=2011, volume=21, issue=3, pageStart=124, pageEnd=130, url=null, language=null, rfNumber=[10], rfOrder=15, authorNames=LI Chao, ZHAO Lindu, journalName=China Safety Science Journal, refType=null, unstructuredReference=LI Chao, ZHAO Lindu. Research on cattle iris localization algorithm and its application in meat food tracking and traceability system[J]. China Safety Science Journal, 2011, 21(3): 124-130. (in Chinese), articleTitle=Research on cattle iris localization algorithm and its application in meat food tracking and traceability system, refAbstract=null), Reference(id=1312006320627216864, tenantId=1146029695717560320, journalId=1311385739772133495, articleId=1312006310527332698, doi=null, pmid=null, pmcid=null, year=2005, volume=18, issue=6, pageStart=868, pageEnd=872, url=null, language=null, rfNumber=[11], rfOrder=16, authorNames=KIM H T, IKEDA Y, CHOI H L, journalName=Asian-Australasian Journal of Animal Sciences, refType=null, unstructuredReference=KIM H T, IKEDA Y, CHOI H L. The identification of Japanese black cattle by their faces[J]. Asian-Australasian Journal of Animal Sciences, 2005, 18(6): 868-872., articleTitle=The identification of Japanese black cattle by their faces, refAbstract=null), Reference(id=1312006320715297249, tenantId=1146029695717560320, journalId=1311385739772133495, articleId=1312006310527332698, doi=null, pmid=null, pmcid=null, year=2013, volume=null, issue=null, pageStart=1, pageEnd=4, url=null, language=null, rfNumber=[12], rfOrder=17, authorNames=CAI C, LI J, journalName=null, refType=null, unstructuredReference=CAI C, LI J. Cattle face recognition using local binary pattern descriptor[C]//2013 Asia-Pacific Signal and Information Processing Association Annual Summit and Conference. IEEE, 2013: 1-4., articleTitle=Cattle face recognition using local binary pattern descriptor, refAbstract=null), Reference(id=1312006320782406114, tenantId=1146029695717560320, journalId=1311385739772133495, articleId=1312006310527332698, doi=null, pmid=null, pmcid=null, year=2017, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=[13], rfOrder=18, authorNames=宋肖肖, journalName=null, refType=null, unstructuredReference=宋肖肖. 牛脸特征点检测的研究与实现[D]. 杨凌: 西北农林科技大学, 2017., articleTitle=牛脸特征点检测的研究与实现, refAbstract=null), Reference(id=1312006320925012451, tenantId=1146029695717560320, journalId=1311385739772133495, articleId=1312006310527332698, doi=null, pmid=null, pmcid=null, year=2017, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=[13], rfOrder=19, authorNames=SONG Xiaoxiao, journalName=null, refType=null, unstructuredReference=SONG Xiaoxiao. Research and implementation of cattle face feature points detection[D]. Yangling: Northwest A&F University, 2017. (in Chinese), articleTitle=Research and implementation of cattle face feature points detection, refAbstract=null), Reference(id=1312006320983732708, tenantId=1146029695717560320, journalId=1311385739772133495, articleId=1312006310527332698, doi=null, pmid=null, pmcid=null, year=2020, volume=null, issue=5, pageStart=59, pageEnd=60, url=null, language=null, rfNumber=[14], rfOrder=20, authorNames=孙曦, 赵平, journalName=大众标准化, refType=null, unstructuredReference=孙曦, 赵平. 基于卷积神经网络的牛脸识别方法研究[J]. 大众标准化, 2020(5): 59-60., articleTitle=基于卷积神经网络的牛脸识别方法研究, refAbstract=null), Reference(id=1312006321063424485, tenantId=1146029695717560320, journalId=1311385739772133495, articleId=1312006310527332698, doi=null, pmid=null, pmcid=null, year=2019, volume=null, issue=13, pageStart=126, pageEnd=null, url=null, language=null, rfNumber=[15], rfOrder=21, authorNames=黄俊华, 田壮, journalName=湖北农机化, refType=null, unstructuredReference=黄俊华, 田壮. 基于改进VGG卷积神经网络的奶牛牛脸识别研究[J]. 湖北农机化, 2019(13): 126., articleTitle=基于改进VGG卷积神经网络的奶牛牛脸识别研究, refAbstract=null), Reference(id=1312006321138921958, tenantId=1146029695717560320, journalId=1311385739772133495, articleId=1312006310527332698, doi=null, pmid=null, pmcid=null, year=2022, volume=42, issue=6, pageStart=40, pageEnd=46, url=null, language=null, rfNumber=[16], rfOrder=22, authorNames=郑鹏, 沈雷, 刘浩, journalName=杭州电子科技大学学报(自然科学版), refType=null, unstructuredReference=郑鹏, 沈雷, 刘浩, 等. 基于Vision Transformer模型的牛脸识别算法[J]. 杭州电子科技大学学报(自然科学版), 2022, 42(6): 40-46., articleTitle=基于Vision Transformer模型的牛脸识别算法, refAbstract=null), Reference(id=1312006321214419431, tenantId=1146029695717560320, journalId=1311385739772133495, articleId=1312006310527332698, doi=null, pmid=null, pmcid=null, year=2023, volume=54, issue=增刊1, pageStart=260, pageEnd=266, url=null, language=null, rfNumber=[17], rfOrder=23, authorNames=刘世锋, 常蕊, 李斌, journalName=农业机械学报, refType=null, unstructuredReference=刘世锋, 常蕊, 李斌, 等. 基于脸部RGB-D图像的牛只个体识别方法[J]. 农业机械学报, 2023, 54(增刊1): 260-266, articleTitle=基于脸部RGB-D图像的牛只个体识别方法, refAbstract=null), Reference(id=1312006321281528296, tenantId=1146029695717560320, journalId=1311385739772133495, articleId=1312006310527332698, doi=null, pmid=null, pmcid=null, year=2023, volume=54, issue=Supp. 1, pageStart=260, pageEnd=266, url=null, language=null, rfNumber=[17], rfOrder=24, authorNames=LIU Shifeng, CHANG Rui, LI Bin, journalName=Transactions of the Chinese Society for Agricultural Machinery, refType=null, unstructuredReference=LIU Shifeng, CHANG Rui, LI Bin, et al. Individual identification of cattle based on RGB-D images[J]. Transactions of the Chinese Society for Agricultural Machinery, 2023, 54(Supp. 1): 260-266. (in Chinese), articleTitle=Individual identification of cattle based on RGB-D images, refAbstract=null), Reference(id=1312006321361220073, tenantId=1146029695717560320, journalId=1311385739772133495, articleId=1312006310527332698, doi=null, pmid=null, pmcid=null, year=2024, volume=null, issue=4, pageStart=36, pageEnd=42, url=null, language=null, rfNumber=[18], rfOrder=25, authorNames=杨胜楠, 赵建敏, 杨梅, journalName=黑龙江畜牧兽医, refType=null, unstructuredReference=杨胜楠, 赵建敏, 杨梅, 等. 基于去偏置项SoftMax和紧致度量损失函数的牛脸识别方法[J]. 黑龙江畜牧兽医, 2024(4): 36-42., articleTitle=基于去偏置项SoftMax和紧致度量损失函数的牛脸识别方法, refAbstract=null), Reference(id=1312006321466077674, tenantId=1146029695717560320, journalId=1311385739772133495, articleId=1312006310527332698, doi=null, pmid=null, pmcid=null, year=2024, volume=null, issue=4, pageStart=36, pageEnd=42, url=null, language=null, rfNumber=[18], rfOrder=26, authorNames=YANG Shengnan, ZHAO Jianmin, YANG Mei, journalName=Heilongjiang Animal Science and Veterinary Medicine, refType=null, unstructuredReference=YANG Shengnan, ZHAO Jianmin, YANG Mei, et al. Cattle face recognition method based on debiased term SoftMax and compact metric loss function[J]. Heilongjiang Animal Science and Veterinary Medicine, 2024(4): 36-42. (in Chinese), articleTitle=Cattle face recognition method based on debiased term SoftMax and compact metric loss function, refAbstract=null), Reference(id=1312006321554158059, tenantId=1146029695717560320, journalId=1311385739772133495, articleId=1312006310527332698, doi=null, pmid=null, pmcid=null, year=2024, volume=null, issue=4, pageStart=43, pageEnd=48, url=null, language=null, rfNumber=[19], rfOrder=27, authorNames=叶孟珂, 李宝山, 杨梅, journalName=黑龙江畜牧兽医, refType=null, unstructuredReference=叶孟珂, 李宝山, 杨梅, 等. 基于深度学习的牛脸识别系统设计[J]. 黑龙江畜牧兽医, 2024(4): 43-48., articleTitle=基于深度学习的牛脸识别系统设计, refAbstract=null), Reference(id=1312006321638044140, tenantId=1146029695717560320, journalId=1311385739772133495, articleId=1312006310527332698, doi=null, pmid=null, pmcid=null, year=2024, volume=null, issue=4, pageStart=43, pageEnd=48, url=null, language=null, rfNumber=[19], rfOrder=28, authorNames=YE Mengke, LI Baoshan, YANG Mei, journalName=Heilongjiang Animal Science and Veterinary Medicine, refType=null, unstructuredReference=YE Mengke, LI Baoshan, YANG Mei, et al. Design of cattle face recognition system based on deep learning[J]. Heilongjiang Animal Science and Veterinary Medicine, 2024(4): 43-48. (in Chinese), articleTitle=Design of cattle face recognition system based on deep learning, refAbstract=null), Reference(id=1312006321721930221, tenantId=1146029695717560320, journalId=1311385739772133495, articleId=1312006310527332698, doi=null, pmid=null, pmcid=null, year=2022, volume=196, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=[20], rfOrder=29, authorNames=WENG Z, MENG F, LIU S, journalName=Computers and Electronics in Agriculture, refType=null, unstructuredReference=WENG Z, MENG F, LIU S, et al. Cattle face recognition based on a two-branch convolutional neural network[J]. Computers and Electronics in Agriculture, 2022, 196: 106871., articleTitle=Cattle face recognition based on a two-branch convolutional neural network, refAbstract=null), Reference(id=1312006321814204910, tenantId=1146029695717560320, journalId=1311385739772133495, articleId=1312006310527332698, doi=null, pmid=null, pmcid=null, year=2024, volume=18, issue=3, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=[21], rfOrder=30, authorNames=BERGMA N, YITZHAKY Y, HALACHMI I, journalName=Animal, refType=null, unstructuredReference=BERGMA N, YITZHAKY Y, HALACHMI I. Biometric identification of dairy cows via real-time facial recognition[J]. Animal, 2024, 18(3): 101079., articleTitle=Biometric identification of dairy cows via real-time facial recognition, refAbstract=null), Reference(id=1312006321889702383, tenantId=1146029695717560320, journalId=1311385739772133495, articleId=1312006310527332698, doi=null, pmid=null, pmcid=null, year=2019, volume=null, issue=null, pageStart=1314, pageEnd=1324, url=null, language=null, rfNumber=[22], rfOrder=31, authorNames=HOWARD A, SANDLER M, CHU G, journalName=null, refType=null, unstructuredReference=HOWARD A, SANDLER M, CHU G, et al. Searching for MobileNetv3[C]//Proceedings of the IEEE/CVF International Conference on Computer Vision, 2019: 1314-1324., articleTitle=Searching for MobileNetv3, refAbstract=null), Reference(id=1312006321956811248, tenantId=1146029695717560320, journalId=1311385739772133495, articleId=1312006310527332698, doi=null, pmid=null, pmcid=null, year=2018, volume=null, issue=null, pageStart=3, pageEnd=19, url=null, language=null, rfNumber=[23], rfOrder=32, authorNames=WOOS, PAPK J, LEE J Y, journalName=null, refType=null, unstructuredReference=WOOS, PAPK J, LEE J Y, et al. Cbam: convolutional block attention module[C]//Proceedings of European Conference on Computer Vision, 2018: 3-19., articleTitle=Cbam: convolutional block attention module, refAbstract=null), Reference(id=1312006322057474545, tenantId=1146029695717560320, journalId=1311385739772133495, articleId=1312006310527332698, doi=null, pmid=null, pmcid=null, year=2013, volume=46, issue=5, pageStart=1415, pageEnd=1432, url=null, language=null, rfNumber=[24], rfOrder=33, authorNames=PRETUZ S, PUIG D, GARCIA M A, journalName=Pattern Recognition, refType=null, unstructuredReference=PRETUZ S, PUIG D, GARCIA M A. Analysis of focus measure operators for shape-from-focus[J]. Pattern Recognition, 2013, 46(5): 1415-1432., articleTitle=Analysis of focus measure operators for shape-from-focus, refAbstract=null), Reference(id=1312006322137166322, tenantId=1146029695717560320, journalId=1311385739772133495, articleId=1312006310527332698, doi=null, pmid=null, pmcid=null, year=2004, volume=13, issue=4, pageStart=600, pageEnd=612, url=null, language=null, rfNumber=[25], rfOrder=34, authorNames=WANG Z, BOVIK A C, SHEIKH H R, journalName=IEEE Transactions on Image Processing, refType=null, unstructuredReference=WANG Z, BOVIK A C, SHEIKH H R, et al. Image quality assessment: from error visibility to structural similarity[J]. IEEE Transactions on Image Processing, 2004, 13(4): 600-612., articleTitle=Image quality assessment: from error visibility to structural similarity, refAbstract=null), Reference(id=1312006322225246707, tenantId=1146029695717560320, journalId=1311385739772133495, articleId=1312006310527332698, doi=null, pmid=null, pmcid=null, year=2015, volume=null, issue=null, pageStart=815, pageEnd=823, url=null, language=null, rfNumber=[26], rfOrder=35, authorNames=SCHROFF F, KALENICHENKO D, PHILBIN J, journalName=null, refType=null, unstructuredReference=SCHROFF F, KALENICHENKO D, PHILBIN J. Facenet: a unified embedding for face recognition and clustering[C]//Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, 2015: 815-823., articleTitle=Facenet: a unified embedding for face recognition and clustering, refAbstract=null), Reference(id=1312006322304938484, tenantId=1146029695717560320, journalId=1311385739772133495, articleId=1312006310527332698, doi=null, pmid=null, pmcid=null, year=2017, volume=31, issue=1, pageStart=4278, pageEnd=4284, url=null, language=null, rfNumber=[27], rfOrder=36, authorNames=SZEGEDY C, IOFFE S, VANHOUCKE V, journalName=null, refType=null, unstructuredReference=SZEGEDY C, IOFFE S, VANHOUCKE V, et al. Inception-v4, inception-resnet and the impact of residual connections on learning[C]//Proceedings of the AAAI Conference on Artificial Intelligence, 2017, 31(1): 4278-4284., articleTitle=Inception-v4, inception-resnet and the impact of residual connections on learning, refAbstract=null), Reference(id=1312006322401407477, tenantId=1146029695717560320, journalId=1311385739772133495, articleId=1312006310527332698, doi=null, pmid=null, pmcid=null, year=2017, volume=null, issue=null, pageStart=5430, pageEnd=5434, url=null, language=null, rfNumber=[28], rfOrder=37, authorNames=BREDIN H, journalName=null, refType=null, unstructuredReference=BREDIN H. TristouNet: triplet loss for speaker turn embedding[C]//Proceedings of the IEEE International Conference on Acoustics, Speech, and Signal Processing (ICASSP). IEEE, 2017: 5430-5434, articleTitle=TristouNet: triplet loss for speaker turn embedding, refAbstract=null), Reference(id=1312006322506265078, tenantId=1146029695717560320, journalId=1311385739772133495, articleId=1312006310527332698, doi=null, pmid=null, pmcid=null, year=2023, volume=54, issue=10, pageStart=275, pageEnd=281, 363, url=null, language=null, rfNumber=[29], rfOrder=38, authorNames=宋怀波, 华志新, 马宝玲, journalName=农业机械学报, refType=null, unstructuredReference=宋怀波, 华志新, 马宝玲, 等. 基于SimCC-ShuffleNetV2的轻量化奶牛关键点检测方法[J]. 农业机械学报, 2023, 54(10): 275-281, 363., articleTitle=基于SimCC-ShuffleNetV2的轻量化奶牛关键点检测方法, refAbstract=null), Reference(id=1312006322573373943, tenantId=1146029695717560320, journalId=1311385739772133495, articleId=1312006310527332698, doi=null, pmid=null, pmcid=null, year=2023, volume=54, issue=10, pageStart=275, pageEnd=281, 363, url=null, language=null, rfNumber=[29], rfOrder=39, authorNames=SONG Huaibo, HUA Zhixin, MA Baoling, journalName=Transactions of the Chinese Society for Agricultural Machinery, refType=null, unstructuredReference=SONG Huaibo, HUA Zhixin, MA Baoling, et al. Lightweight keypoint detection method of dairy cow based on SimCC-ShuffleNetV2[J]. Transactions of the Chinese Society for Agricultural Machinery, 2023, 54(10): 275-281, 363. (in Chinese), articleTitle=Lightweight keypoint detection method of dairy cow based on SimCC-ShuffleNetV2, refAbstract=null), Reference(id=1312006322640482808, tenantId=1146029695717560320, journalId=1311385739772133495, articleId=1312006310527332698, doi=null, pmid=null, pmcid=null, year=2018, volume=null, issue=null, pageStart=4510, pageEnd=4520, url=null, language=null, rfNumber=[30], rfOrder=40, authorNames=SANDLER M, HOWARD A, ZHU M, journalName=null, refType=null, unstructuredReference=SANDLER M, HOWARD A, ZHU M, et al. Mobilenetv2: inverted residuals and linear bottlenecks[C]//Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, 2018: 4510-4520., articleTitle=Mobilenetv2: inverted residuals and linear bottlenecks, refAbstract=null), Reference(id=1312006322711785977, tenantId=1146029695717560320, journalId=1311385739772133495, articleId=1312006310527332698, doi=null, pmid=null, pmcid=null, year=2015, volume=null, issue=null, pageStart=730, pageEnd=734, url=null, language=null, rfNumber=[31], rfOrder=41, authorNames=LIU S, DENG W, journalName=null, refType=null, unstructuredReference=LIU S, DENG W. Very deep convolutional neuralnetwork based image classification using small training sample size[C]//2015 3rd IAPR Asian Conference on Pattern Recognition (ACPR). IEEE, 2015: 730-734., articleTitle=Very deep convolutional neuralnetwork based image classification using small training sample size, refAbstract=null), Reference(id=1312006322770506234, tenantId=1146029695717560320, journalId=1311385739772133495, articleId=1312006310527332698, doi=null, pmid=null, pmcid=null, year=2022, volume=45, issue=1, pageStart=87, pageEnd=110, url=null, language=null, rfNumber=[32], rfOrder=42, authorNames=HAN K, WANG Y, CHEN H, journalName=IEEE Transactions on Pattern Analysis and Machine Intelligence, refType=null, unstructuredReference=HAN K, WANG Y, CHEN H, et al. A survey on vision transformer[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2022, 45(1): 87-110., articleTitle=A survey on vision transformer, refAbstract=null), Reference(id=1312006322837615099, tenantId=1146029695717560320, journalId=1311385739772133495, articleId=1312006310527332698, doi=null, pmid=null, pmcid=null, year=2021, volume=null, issue=null, pageStart=10012, pageEnd=10022, url=null, language=null, rfNumber=[33], rfOrder=43, authorNames=LIU Z, LIN Y, CAO Y, journalName=null, refType=null, unstructuredReference=LIU Z, LIN Y, CAO Y, et al. Swin transformer: hierarchical vision transformer using shifted windows[C]//Proceedings of the IEEE/CVF International Conference on Computer Vision, 2021: 10012-10022., articleTitle=Swin transformer: hierarchical vision transformer using shifted windows, refAbstract=null), Reference(id=1312006322904723964, tenantId=1146029695717560320, journalId=1311385739772133495, articleId=1312006310527332698, doi=null, pmid=null, pmcid=null, year=2016, volume=null, issue=null, pageStart=770, pageEnd=778, url=null, language=null, rfNumber=[34], rfOrder=44, authorNames=HE K, ZHANG X, REN S, journalName=null, refType=null, unstructuredReference=HE K, ZHANG X, REN S, et al. Deep residual learning for image recognition[C]//Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, 2016: 770-778., articleTitle=Deep residual learning for image recognition, refAbstract=null)], funds=[Fund(id=1312006319276650959, tenantId=1146029695717560320, journalId=1311385739772133495, articleId=1312006310527332698, awardId=20180201069GX, language=CN, fundingSource=吉林省重点科技研发项目(20180201069GX), fundOrder=null, country=null)], companyList=[AuthorCompany(id=1312006314411258243, tenantId=1146029695717560320, journalId=1311385739772133495, articleId=1312006310527332698, xref=1., ext=[AuthorCompanyExt(id=1312006314419646852, tenantId=1146029695717560320, journalId=1311385739772133495, articleId=1312006310527332698, companyId=1312006314411258243, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1.School of Computer Science and Technology, Changchun University of Science and Technology, Changchun 130012, China), AuthorCompanyExt(id=1312006314423841157, tenantId=1146029695717560320, journalId=1311385739772133495, articleId=1312006310527332698, companyId=1312006314411258243, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1.长春理工大学计算机科学技术学院,长春 130012)]), AuthorCompany(id=1312006314499338630, tenantId=1146029695717560320, journalId=1311385739772133495, articleId=1312006310527332698, xref=2., ext=[AuthorCompanyExt(id=1312006314503532935, tenantId=1146029695717560320, journalId=1311385739772133495, articleId=1312006310527332698, companyId=1312006314499338630, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=2.Jilin Sino Agri Sunshine Data Co., Ltd., Changchun 130022, China), AuthorCompanyExt(id=1312006314511921544, tenantId=1146029695717560320, journalId=1311385739772133495, articleId=1312006310527332698, companyId=1312006314499338630, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=2.吉林省中农阳光数据有限公司,长春 130022)])], figs=[ArticleFig(id=1312006316651016623, tenantId=1146029695717560320, journalId=1311385739772133495, articleId=1312006310527332698, language=EN, label=Fig. 1, caption=Cattle face data collection environment in cattle farms, figureFileSmall=UGkXBZU17Wsd5PeQu3UM/g==, figureFileBig=1ztwDc9//k58LPtrjgIl8w==, tableContent=null), ArticleFig(id=1312006316730708400, tenantId=1146029695717560320, journalId=1311385739772133495, articleId=1312006310527332698, language=CN, label=图1, caption=牛场牛脸数据采集环境, figureFileSmall=UGkXBZU17Wsd5PeQu3UM/g==, figureFileBig=1ztwDc9//k58LPtrjgIl8w==, tableContent=null), ArticleFig(id=1312006316932034993, tenantId=1146029695717560320, journalId=1311385739772133495, articleId=1312006310527332698, language=EN, label=Fig. 2, caption=Annotated key points on frontal cattle face diagram, figureFileSmall=vqi7etJntq+5bmF+eGNiiw==, figureFileBig=wBdayEfgLm1TC9vIddxHtQ==, tableContent=null), ArticleFig(id=1312006317024309682, tenantId=1146029695717560320, journalId=1311385739772133495, articleId=1312006310527332698, language=CN, label=图2, caption=牛头正脸关键点标注位置图, figureFileSmall=vqi7etJntq+5bmF+eGNiiw==, figureFileBig=wBdayEfgLm1TC9vIddxHtQ==, tableContent=null), ArticleFig(id=1312006317083029939, tenantId=1146029695717560320, journalId=1311385739772133495, articleId=1312006310527332698, language=EN, label=Fig. 3, caption=Improved YOLO v8 model, figureFileSmall=rcV4OuivP5bpqCzrz5NQ/g==, figureFileBig=qZObJcvM135S1HOvcQCQfw==, tableContent=null), ArticleFig(id=1312006317150138804, tenantId=1146029695717560320, journalId=1311385739772133495, articleId=1312006310527332698, language=CN, label=图3, caption=改进的YOLO v8模型, figureFileSmall=rcV4OuivP5bpqCzrz5NQ/g==, figureFileBig=qZObJcvM135S1HOvcQCQfw==, tableContent=null), ArticleFig(id=1312006317213053365, tenantId=1146029695717560320, journalId=1311385739772133495, articleId=1312006310527332698, language=EN, label=Fig. 4, caption=Cattle face cropping dataset, figureFileSmall=eRYdTO+lPSHNFt79r9k7aw==, figureFileBig=xfrjMiQ8xlZGyqm5GNNZ0g==, tableContent=null), ArticleFig(id=1312006317280162230, tenantId=1146029695717560320, journalId=1311385739772133495, articleId=1312006310527332698, language=CN, label=图4, caption=牛脸裁剪数据集, figureFileSmall=eRYdTO+lPSHNFt79r9k7aw==, figureFileBig=xfrjMiQ8xlZGyqm5GNNZ0g==, tableContent=null), ArticleFig(id=1312006317364048311, tenantId=1146029695717560320, journalId=1311385739772133495, articleId=1312006310527332698, language=EN, label=Fig. 5, caption=Framework for cattle face recognition, figureFileSmall=+CH4Eh07BDV5YWtWz7w9Lg==, figureFileBig=nVVjPYI6d08avV1wJHPp0Q==, tableContent=null), ArticleFig(id=1312006317431157176, tenantId=1146029695717560320, journalId=1311385739772133495, articleId=1312006310527332698, language=CN, label=图5, caption=牛脸识别框架, figureFileSmall=+CH4Eh07BDV5YWtWz7w9Lg==, figureFileBig=nVVjPYI6d08avV1wJHPp0Q==, tableContent=null), ArticleFig(id=1312006317510848953, tenantId=1146029695717560320, journalId=1311385739772133495, articleId=1312006310527332698, language=EN, label=Fig. 6, caption=Improved Inception-ResNet network architecture, figureFileSmall=WLYpTtL1xaVCMVyiwITwNw==, figureFileBig=Tww53j2N05pigjPcCf1WGA==, tableContent=null), ArticleFig(id=1312006317594735034, tenantId=1146029695717560320, journalId=1311385739772133495, articleId=1312006310527332698, language=CN, label=图6, caption=改进Inception-ResNet网络结构, figureFileSmall=WLYpTtL1xaVCMVyiwITwNw==, figureFileBig=Tww53j2N05pigjPcCf1WGA==, tableContent=null), ArticleFig(id=1312006317666038203, tenantId=1146029695717560320, journalId=1311385739772133495, articleId=1312006310527332698, language=EN, label=Fig. 7, caption=Triplet loss schematic, figureFileSmall=h/CCQ14KcD2WBIheaIIUpw==, figureFileBig=NJ9mHoTRV1b7CW06oEi92w==, tableContent=null), ArticleFig(id=1312006317737341372, tenantId=1146029695717560320, journalId=1311385739772133495, articleId=1312006310527332698, language=CN, label=图7, caption=三元组损失示意图, figureFileSmall=h/CCQ14KcD2WBIheaIIUpw==, figureFileBig=NJ9mHoTRV1b7CW06oEi92w==, tableContent=null), ArticleFig(id=1312006317825421757, tenantId=1146029695717560320, journalId=1311385739772133495, articleId=1312006310527332698, language=EN, label=Fig. 8, caption=Actual cattle face occlusion image with cow face occlusion dataset, figureFileSmall=S5b+7QF082xrPBsdFVCcgg==, figureFileBig=qcWKOWM4Hcog5tk+DCmF5Q==, tableContent=null), ArticleFig(id=1312006317913502142, tenantId=1146029695717560320, journalId=1311385739772133495, articleId=1312006310527332698, language=CN, label=图8, caption=实际牛脸遮挡图像与牛脸遮挡数据集, figureFileSmall=S5b+7QF082xrPBsdFVCcgg==, figureFileBig=qcWKOWM4Hcog5tk+DCmF5Q==, tableContent=null), ArticleFig(id=1312006317988999615, tenantId=1146029695717560320, journalId=1311385739772133495, articleId=1312006310527332698, language=EN, label=Fig. 9, caption=Recognition performance in different environments and for various cattle breeds, figureFileSmall=QFdEymRcbYc5oLGRm0NS1w==, figureFileBig=fzEBaD75gFrFtYA/HteWPw==, tableContent=null), ArticleFig(id=1312006318106440128, tenantId=1146029695717560320, journalId=1311385739772133495, articleId=1312006310527332698, language=CN, label=图9, caption=不同场景和不同品种下的识别效果, figureFileSmall=QFdEymRcbYc5oLGRm0NS1w==, figureFileBig=fzEBaD75gFrFtYA/HteWPw==, tableContent=null), ArticleFig(id=1312006318186131905, tenantId=1146029695717560320, journalId=1311385739772133495, articleId=1312006310527332698, language=EN, label=Tab. 1, caption=

Comparison of cattle face keypoint detection results

, figureFileSmall=null, figureFileBig=null, tableContent=
模型数据集NME/%推理时间/ms参数量牛脸识别任务准确率/%
YOLO v8CowK3.11254.03×10695.9
本文CowK3.52101.95×10695.8
), ArticleFig(id=1312006318261629378, tenantId=1146029695717560320, journalId=1311385739772133495, articleId=1312006310527332698, language=CN, label=表1, caption=

牛脸关键点检测对比结果

, figureFileSmall=null, figureFileBig=null, tableContent=
模型数据集NME/%推理时间/ms参数量牛脸识别任务准确率/%
YOLO v8CowK3.11254.03×10695.9
本文CowK3.52101.95×10695.8
), ArticleFig(id=1312006318349709763, tenantId=1146029695717560320, journalId=1311385739772133495, articleId=1312006310527332698, language=EN, label=Tab. 2, caption=

Comparison results of cattle face recognition under keypoint cropping and correction techniques

, figureFileSmall=null, figureFileBig=null, tableContent=
类型评估数据集准确率精确率召回率假正率F1分数
未使用裁剪CowTc+CowF79.580.291.845.285.6
29个关键点裁剪CowTc+CowF87.988.393.923.791.0
43个关键点裁剪CowTc+CowF95.897.296.55.796.8
), ArticleFig(id=1312006318433595844, tenantId=1146029695717560320, journalId=1311385739772133495, articleId=1312006310527332698, language=CN, label=表2, caption=

关键点裁剪矫正技术下的牛脸识别对比结果

, figureFileSmall=null, figureFileBig=null, tableContent=
类型评估数据集准确率精确率召回率假正率F1分数
未使用裁剪CowTc+CowF79.580.291.845.285.6
29个关键点裁剪CowTc+CowF87.988.393.923.791.0
43个关键点裁剪CowTc+CowF95.897.296.55.796.8
), ArticleFig(id=1312006318521676229, tenantId=1146029695717560320, journalId=1311385739772133495, articleId=1312006310527332698, language=EN, label=Tab. 3, caption=

Comparison results of cattle face recognition with different network models

, figureFileSmall=null, figureFileBig=null, tableContent=
网络模型评估数据集准确率精确率召回率假正率F1分数
MobileNetv2CowTc+CowF91.394.991.910.093.4
Inception-ResNetv2CowTc+CowF91.892.895.114.893.9
VGG16CowTc+CowF86.789.990.220.590.0
Vision-TransfomerCowTc+CowF90.794.092.011.893.0
Swin-TransfomerCowTc+CowF94.098.492.53.095.4
ResNet50CowTc+CowF91.195.191.39.593.2
本文模型CowTc+CowF95.897.296.55.796.8
), ArticleFig(id=1312006318588785094, tenantId=1146029695717560320, journalId=1311385739772133495, articleId=1312006310527332698, language=CN, label=表3, caption=

不同网络模型牛脸识别对比结果

, figureFileSmall=null, figureFileBig=null, tableContent=
网络模型评估数据集准确率精确率召回率假正率F1分数
MobileNetv2CowTc+CowF91.394.991.910.093.4
Inception-ResNetv2CowTc+CowF91.892.895.114.893.9
VGG16CowTc+CowF86.789.990.220.590.0
Vision-TransfomerCowTc+CowF90.794.092.011.893.0
Swin-TransfomerCowTc+CowF94.098.492.53.095.4
ResNet50CowTc+CowF91.195.191.39.593.2
本文模型CowTc+CowF95.897.296.55.796.8
), ArticleFig(id=1312006318647505351, tenantId=1146029695717560320, journalId=1311385739772133495, articleId=1312006310527332698, language=EN, label=Tab. 4, caption=

Comparison of accuracy of cow face recognition with addition of new cattle data%

, figureFileSmall=null, figureFileBig=null, tableContent=
网络模型测试数据集原模型准确率50幅图像100幅图像250幅图像350幅图像600幅图像
MobileNetv2CowTc91.991.991.791.791.791.7
Inception-ResNetv2CowTc94.494.294.194.093.993.8
VGG16CowTc89.589.589.589.589.589.5
Vision-TransfomerCowTc91.891.591.591.491.291.1
Swin-TransfomerCowTc92.492.492.191.891.791.5
ResNet50CowTc90.990.990.990.990.990.9
本文模型CowTc96.496.295.895.795.495.3
), ArticleFig(id=1312006318723002824, tenantId=1146029695717560320, journalId=1311385739772133495, articleId=1312006310527332698, language=CN, label=表4, caption=

牛脸识别在添加新增牛只数据下准确率对比

, figureFileSmall=null, figureFileBig=null, tableContent=
网络模型测试数据集原模型准确率50幅图像100幅图像250幅图像350幅图像600幅图像
MobileNetv2CowTc91.991.991.791.791.791.7
Inception-ResNetv2CowTc94.494.294.194.093.993.8
VGG16CowTc89.589.589.589.589.589.5
Vision-TransfomerCowTc91.891.591.591.491.291.1
Swin-TransfomerCowTc92.492.492.191.891.791.5
ResNet50CowTc90.990.990.990.990.990.9
本文模型CowTc96.496.295.895.795.495.3
), ArticleFig(id=1312006318785917385, tenantId=1146029695717560320, journalId=1311385739772133495, articleId=1312006310527332698, language=EN, label=Tab. 5, caption=

Comparison results of ablation experiments for cattle face recognition models

, figureFileSmall=null, figureFileBig=null, tableContent=
Inception-ResNetv2DSConvCBAM准确率/%精确率/%召回率/%假正率/%F1分数/%参数量推理时间/ms
√××90.992.194.516.393.32.26×10757
√√×91.091.395.618.593.41.54×10735
√√√95.897.296.55.796.81.95×10739
), ArticleFig(id=1312006318869803466, tenantId=1146029695717560320, journalId=1311385739772133495, articleId=1312006310527332698, language=CN, label=表5, caption=

牛脸识别模型消融实验对比结果

, figureFileSmall=null, figureFileBig=null, tableContent=
Inception-ResNetv2DSConvCBAM准确率/%精确率/%召回率/%假正率/%F1分数/%参数量推理时间/ms
√××90.992.194.516.393.32.26×10757
√√×91.091.395.618.593.41.54×10735
√√√95.897.296.55.796.81.95×10739
), ArticleFig(id=1312006318941106635, tenantId=1146029695717560320, journalId=1311385739772133495, articleId=1312006310527332698, language=EN, label=Tab. 6, caption=

Cattle face recognition performance evaluation

, figureFileSmall=null, figureFileBig=null, tableContent=
牛脸检测模型参数量检测推理时间/ms牛脸识别模型参数量识别推理时间/ms总计推理时间/ms
YOLO v84.03×10625MobileNetv23.21×1062247
Inception-ResNetv23.04×1073560
VGG161.47×1073964
Vision-Transfomer8.56×1074368
Swin-Transfomer2.75×107135160
ResNet502.35×1074873
本文模型1.95×1072954
本文模型1.95×10610MobileNetv23.21×1062232
Inception-ResNetv23.04×1073545
VGG161.47×1073949
Vision-Transfomer8.56×1074353
Swin-Transfomer2.75×107135145
ResNet502.35×1074858
本文模型1.95×1072939
), ArticleFig(id=1312006319004021196, tenantId=1146029695717560320, journalId=1311385739772133495, articleId=1312006310527332698, language=CN, label=表6, caption=

牛脸识别性能评估

, figureFileSmall=null, figureFileBig=null, tableContent=
牛脸检测模型参数量检测推理时间/ms牛脸识别模型参数量识别推理时间/ms总计推理时间/ms
YOLO v84.03×10625MobileNetv23.21×1062247
Inception-ResNetv23.04×1073560
VGG161.47×1073964
Vision-Transfomer8.56×1074368
Swin-Transfomer2.75×107135160
ResNet502.35×1074873
本文模型1.95×1072954
本文模型1.95×10610MobileNetv23.21×1062232
Inception-ResNetv23.04×1073545
VGG161.47×1073949
Vision-Transfomer8.56×1074353
Swin-Transfomer2.75×107135145
ResNet502.35×1074858
本文模型1.95×1072939
), ArticleFig(id=1312006319083712973, tenantId=1146029695717560320, journalId=1311385739772133495, articleId=1312006310527332698, language=EN, label=Tab. 7, caption=

Average precision of cattle face keypoint detection under different occlusions

, figureFileSmall=null, figureFileBig=null, tableContent=
牛脸遮挡类别遮挡比例/%
051015202550
内部单个区域遮挡95.794.391.490.985.781.541.7
内部多个区域遮挡95.794.392.991.485.183.747.8
外部区域遮挡95.795.795.795.794.392.782.9
), ArticleFig(id=1312006319138238926, tenantId=1146029695717560320, journalId=1311385739772133495, articleId=1312006310527332698, language=CN, label=表7, caption=

不同遮挡情况下牛脸关键点平均精度

, figureFileSmall=null, figureFileBig=null, tableContent=
牛脸遮挡类别遮挡比例/%
051015202550
内部单个区域遮挡95.794.391.490.985.781.541.7
内部多个区域遮挡95.794.392.991.485.183.747.8
外部区域遮挡95.795.795.795.794.392.782.9
)], attaches=null, journal=Journal(id=1311384913490051186, delFlag=0, nameCn=农业机械学报, nameEn=Transactions of the Chinese Society for Agricultural Machinery, nameHistory1=null, nameHistory2=null, issn=1000-1298, eissn=null, cn=11-1964/S, coden=CODEN NUYCA3, periodic=0, 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=DpMljKbCTPO4JEohCaKm5A==, journalPrice=null, startedYear=null, abbrevIsoEn=Transactions of the Chinese Society for Agricultural Machinery, journalRemark=null, publicationField=null, createdTime=1790588127731, updatedTime=1790740716593, createdBy=18614031015, updatedBy=18614031015, firstLetterCn=N, firstLetterEn=N, subjectCode=Agriculture, subjectName=null, subjectCodeEn=Agriculture, subjectNameEn=null, picCn=DpMljKbCTPO4JEohCaKm5A==, picEn=klDuz9nOYH3sS2mMFZcDKQ==, jcr=null, cjcr=null, exts=[JournalExt(id=1311386928897945885, language=CN, name=农业机械学报, 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=1790588608240, updatedTime=1790588608240, createdBy=13701087609, updatedBy=13701087609, submissionGuidelinesUrl=, submissionAuthorUrl=https://www.j-csam.org/jcsam/author/login, submissionEditorUrl=https://www.j-csam.org/jcsam/editor/login, submissionReviewUrl=https://www.j-csam.org/jcsam/user/login, submissionCeEditorUrl=, submissionAeEditorUrl=, option={"copyright":""}), JournalExt(id=1311386928944083230, language=EN, name=Transactions of the Chinese Society of Agricultural Machinery, 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=1790588608251, updatedTime=1790588608251, createdBy=13701087609, updatedBy=13701087609, submissionGuidelinesUrl=, submissionAuthorUrl=https://www.j-csam.org/jcsam/author/login, submissionEditorUrl=https://www.j-csam.org/jcsam/editor/login, submissionReviewUrl=https://www.j-csam.org/jcsam/user/login, submissionCeEditorUrl=, submissionAeEditorUrl=, option={"copyright":""})], databaseList=null, tenantJournalId=1311385739772133495, websiteList=[Website(id=1311386135197221186, webName=null, webTitle=null, webDomain=null, webCopyrigh=null, webIpcNo=null, seoTitle=null, seoKeywords=null, seoDescription=null, tenantJournalId=null, journalId=1311385739772133495, 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/nyjxxb/CN, language=CN, createTime=1790588419007, createBy=18614031015, updateTime=1790588455647, updateBy=18614031015, name=农业机械学报-中文, tplId=1146099689490845704, title=农业机械学报, delFlag=0, indexPage=/home, props=[WebsiteProps(id=1312024622120399707, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1311386135197221186, code=articleTextType, value=kx, createTime=1790740646152, updateTime=1790740646152, creator=18614031015, updator=18614031015), WebsiteProps(id=1312024622091039576, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1311386135197221186, code=banner, value=null, createTime=1790740646145, updateTime=1790740646145, creator=18614031015, updator=18614031015), WebsiteProps(id=1312024622149759838, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1311386135197221186, code=grayFlag, value=0, createTime=1790740646159, updateTime=1790740646159, creator=18614031015, updator=18614031015), WebsiteProps(id=1312024622078456663, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1311386135197221186, code=logo, value=https://castjournals.cast.org.cn/joweb/nyjxxb/CN/file/pic?fileId=oEf7LShi6rv3EG1pNx7PwQ==, createTime=1790740646142, updateTime=1790740646142, creator=18614031015, updator=18614031015), WebsiteProps(id=1312024622166537056, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1311386135197221186, code=minRunFlag, value=0, createTime=1790740646163, updateTime=1790740646163, creator=18614031015, updator=18614031015), WebsiteProps(id=1312024622112011098, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1311386135197221186, code=picServerUrl, value=https://castjournals.cast.org.cn/joweb/nyjxxb/CN/file/pic, createTime=1790740646150, updateTime=1790740646150, creator=18614031015, updator=18614031015), WebsiteProps(id=1312024622158148447, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1311386135197221186, code=silenceFlag, value=0, createTime=1790740646161, updateTime=1790740646161, creator=18614031015, updator=18614031015), WebsiteProps(id=1312024622099428185, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1311386135197221186, code=staticResourcePath, value=https://castjournals.cast.org.cn/joweb/cast_kjdb_cn_619/, createTime=1790740646147, updateTime=1790740646147, creator=18614031015, updator=18614031015), WebsiteProps(id=1312024622128788316, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1311386135197221186, code=themeColor, value=null, createTime=1790740646154, updateTime=1790740646154, creator=18614031015, updator=18614031015), WebsiteProps(id=1312024622141371229, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1311386135197221186, code=themeStyle, value=null, createTime=1790740646157, updateTime=1790740646157, creator=18614031015, updator=18614031015)]), Website(id=1311386135398547788, webName=null, webTitle=null, webDomain=null, webCopyrigh=null, webIpcNo=null, seoTitle=null, seoKeywords=null, seoDescription=null, tenantJournalId=null, journalId=1311385739772133495, 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/nyjxxb/EN, language=EN, createTime=1790588419055, createBy=18614031015, updateTime=1790740692575, updateBy=18614031015, name=农业机械学报-英文, tplId=1146101810881728533, title=Transactions of the Chinese Society for Agricultural Machinery, delFlag=0, indexPage=/home, props=[WebsiteProps(id=1312024677325828967, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1311386135398547788, code=articleTextType, value=kx, createTime=1790740659314, updateTime=1790740659314, creator=18614031015, updator=18614031015), WebsiteProps(id=1312024677300663140, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1311386135398547788, code=banner, value=null, createTime=1790740659308, updateTime=1790740659308, creator=18614031015, updator=18614031015), WebsiteProps(id=1312024677350994794, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1311386135398547788, code=grayFlag, value=0, createTime=1790740659320, updateTime=1790740659320, creator=18614031015, updator=18614031015), WebsiteProps(id=1312024677288080227, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1311386135398547788, code=logo, value=https://castjournals.cast.org.cn/joweb/nyjxxb/EN/file/pic?fileId=oEf7LShi6rv3EG1pNx7PwQ==, createTime=1790740659305, updateTime=1790740659305, creator=18614031015, updator=18614031015), WebsiteProps(id=1312024677367772012, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1311386135398547788, code=minRunFlag, value=0, createTime=1790740659324, updateTime=1790740659324, creator=18614031015, updator=18614031015), WebsiteProps(id=1312024677317440358, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1311386135398547788, code=picServerUrl, value=https://castjournals.cast.org.cn/joweb/nyjxxb/EN/file/pic, createTime=1790740659312, updateTime=1790740659312, creator=18614031015, updator=18614031015), WebsiteProps(id=1312024677359383403, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1311386135398547788, code=silenceFlag, value=0, createTime=1790740659322, updateTime=1790740659322, creator=18614031015, updator=18614031015), WebsiteProps(id=1312024677309051749, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1311386135398547788, code=staticResourcePath, value=https://castjournals.cast.org.cn/joweb/cast_kjdb_en_623/, createTime=1790740659310, updateTime=1790740659310, creator=18614031015, updator=18614031015), WebsiteProps(id=1312024677334217576, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1311386135398547788, code=themeColor, value=null, createTime=1790740659316, updateTime=1790740659316, creator=18614031015, updator=18614031015), WebsiteProps(id=1312024677342606185, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1311386135398547788, code=themeStyle, value=null, createTime=1790740659318, updateTime=1790740659318, creator=18614031015, updator=18614031015)])], journalTitle=农业机械学报, weixinUrl=null, journalUrl=https://www.j-csam.org/, iacademicId=null, status=1, seqNo=null, journalTitleEn=Transactions of the Chinese Society for Agricultural Machinery, journalPhotoCn=DpMljKbCTPO4JEohCaKm5A==, journalPhotoEn=klDuz9nOYH3sS2mMFZcDKQ==, journalFirstLetter=N, 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/nyjxxb/CN/10.6041/j.issn.1000-1298.2025.12.054, detailUrlEn=https://castjournals.cast.org.cn/joweb/nyjxxb/EN/10.6041/j.issn.1000-1298.2025.12.054, pdfUrlCn=https://castjournals.cast.org.cn/joweb/nyjxxb/CN/PDF/10.6041/j.issn.1000-1298.2025.12.054, pdfUrlEn=https://castjournals.cast.org.cn/joweb/nyjxxb/EN/PDF/10.6041/j.issn.1000-1298.2025.12.054, aliStartDate=0, aliEndDate=0, collectionFlag=false, citedCount=null, citedUrl=null, previewStatus=0, delFlag=0, hasFullText=1, orderTime=1766592000000, fullTextJson=null, articleText=null, reference=null)
收藏切换
基于改进YOLO v8关键点检测的西门塔尔牛只个体识别方法
收藏切换
PDF下载
田振江 1 , 韩成 1 , 张超 1 , 李月莹 1 , 于福东 2 , 钟馨慧 2
农业机械学报 | 农业信息化工程 2025,56(12): 591-602
收起
收藏切换
农业机械学报 |农业信息化工程 2025 , 56 (12) : 591 -602
基于改进YOLO v8关键点检测的西门塔尔牛只个体识别方法
全屏
[Author(id=1312006314591613322, tenantId=1146029695717560320, journalId=1311385739772133495, articleId=1312006310527332698, orderNo=0, firstName=null, middleName=null, lastName=null, nameCn=null, orcid=null, stid=null, country=null, authorPic=null, dead=0, email=2022101001@mails.cust.edu.cn, emailSecond=null, emailThird=null, correspondingAuthor=0, authorType=1, ext={EN=AuthorExt(id=1312006314679693708, tenantId=1146029695717560320, journalId=1311385739772133495, articleId=1312006310527332698, authorId=1312006314591613322, language=EN, stringName=Zhenjiang TIAN, firstName=Zhenjiang, middleName=null, lastName=TIAN, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, address=1.School of Computer Science and Technology, Changchun University of Science and Technology, Changchun 130012, China, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null), CN=AuthorExt(id=1312006314776162701, tenantId=1146029695717560320, journalId=1311385739772133495, articleId=1312006310527332698, authorId=1312006314591613322, language=CN, stringName=田振江, firstName=null, middleName=null, lastName=null, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, address=1.长春理工大学计算机科学技术学院,长春 130012, bio={"content":"

田振江(1998—),男,博士生,主要从事牛脸面部识别研究,E-mail:

"}, bioImg=null, bioContent=

田振江(1998—),男,博士生,主要从事牛脸面部识别研究,E-mail:

, aboutCorrespAuthor=null)}, companyList=[AuthorCompany(id=1312006314411258243, tenantId=1146029695717560320, journalId=1311385739772133495, articleId=1312006310527332698, xref=1., ext=[AuthorCompanyExt(id=1312006314419646852, tenantId=1146029695717560320, journalId=1311385739772133495, articleId=1312006310527332698, companyId=1312006314411258243, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1.School of Computer Science and Technology, Changchun University of Science and Technology, Changchun 130012, China), AuthorCompanyExt(id=1312006314423841157, tenantId=1146029695717560320, journalId=1311385739772133495, articleId=1312006310527332698, companyId=1312006314411258243, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1.长春理工大学计算机科学技术学院,长春 130012)])]), Author(id=1312006314847465871, tenantId=1146029695717560320, journalId=1311385739772133495, articleId=1312006310527332698, orderNo=1, firstName=null, middleName=null, lastName=null, nameCn=null, orcid=null, stid=null, country=null, authorPic=null, dead=0, email=hancheng@cust.edu.cn, emailSecond=null, emailThird=null, correspondingAuthor=1, authorType=1, ext={EN=AuthorExt(id=1312006314927157649, tenantId=1146029695717560320, journalId=1311385739772133495, articleId=1312006310527332698, authorId=1312006314847465871, language=EN, stringName=Cheng HAN, firstName=Cheng, middleName=null, lastName=HAN, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, address=1.School of Computer Science and Technology, Changchun University of Science and Technology, Changchun 130012, China, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null), CN=AuthorExt(id=1312006314998460818, tenantId=1146029695717560320, journalId=1311385739772133495, articleId=1312006310527332698, authorId=1312006314847465871, language=CN, stringName=韩成, firstName=null, middleName=null, lastName=null, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, address=1.长春理工大学计算机科学技术学院,长春 130012, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null)}, companyList=[AuthorCompany(id=1312006314411258243, tenantId=1146029695717560320, journalId=1311385739772133495, articleId=1312006310527332698, xref=1., ext=[AuthorCompanyExt(id=1312006314419646852, tenantId=1146029695717560320, journalId=1311385739772133495, articleId=1312006310527332698, companyId=1312006314411258243, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1.School of Computer Science and Technology, Changchun University of Science and Technology, Changchun 130012, China), AuthorCompanyExt(id=1312006314423841157, tenantId=1146029695717560320, journalId=1311385739772133495, articleId=1312006310527332698, companyId=1312006314411258243, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1.长春理工大学计算机科学技术学院,长春 130012)])]), Author(id=1312006315061375380, tenantId=1146029695717560320, journalId=1311385739772133495, articleId=1312006310527332698, 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=1312006315132678550, tenantId=1146029695717560320, journalId=1311385739772133495, articleId=1312006310527332698, authorId=1312006315061375380, language=EN, stringName=Chao ZHANG, firstName=Chao, middleName=null, lastName=ZHANG, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, address=1.School of Computer Science and Technology, Changchun University of Science and Technology, Changchun 130012, China, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null), CN=AuthorExt(id=1312006315220758935, tenantId=1146029695717560320, journalId=1311385739772133495, articleId=1312006310527332698, authorId=1312006315061375380, language=CN, stringName=张超, firstName=null, middleName=null, lastName=null, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, address=1.长春理工大学计算机科学技术学院,长春 130012, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null)}, companyList=[AuthorCompany(id=1312006314411258243, tenantId=1146029695717560320, journalId=1311385739772133495, articleId=1312006310527332698, xref=1., ext=[AuthorCompanyExt(id=1312006314419646852, tenantId=1146029695717560320, journalId=1311385739772133495, articleId=1312006310527332698, companyId=1312006314411258243, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1.School of Computer Science and Technology, Changchun University of Science and Technology, Changchun 130012, China), AuthorCompanyExt(id=1312006314423841157, tenantId=1146029695717560320, journalId=1311385739772133495, articleId=1312006310527332698, companyId=1312006314411258243, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1.长春理工大学计算机科学技术学院,长春 130012)])]), Author(id=1312006315283673497, tenantId=1146029695717560320, journalId=1311385739772133495, articleId=1312006310527332698, 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=1312006315367559579, tenantId=1146029695717560320, journalId=1311385739772133495, articleId=1312006310527332698, authorId=1312006315283673497, language=EN, stringName=Yueying LI, firstName=Yueying, middleName=null, lastName=LI, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, address=1.School of Computer Science and Technology, Changchun University of Science and Technology, Changchun 130012, China, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null), CN=AuthorExt(id=1312006315438862748, tenantId=1146029695717560320, journalId=1311385739772133495, articleId=1312006310527332698, authorId=1312006315283673497, language=CN, stringName=李月莹, firstName=null, middleName=null, lastName=null, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, address=1.长春理工大学计算机科学技术学院,长春 130012, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null)}, companyList=[AuthorCompany(id=1312006314411258243, tenantId=1146029695717560320, journalId=1311385739772133495, articleId=1312006310527332698, xref=1., ext=[AuthorCompanyExt(id=1312006314419646852, tenantId=1146029695717560320, journalId=1311385739772133495, articleId=1312006310527332698, companyId=1312006314411258243, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1.School of Computer Science and Technology, Changchun University of Science and Technology, Changchun 130012, China), AuthorCompanyExt(id=1312006314423841157, tenantId=1146029695717560320, journalId=1311385739772133495, articleId=1312006310527332698, companyId=1312006314411258243, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1.长春理工大学计算机科学技术学院,长春 130012)])]), Author(id=1312006315505971614, tenantId=1146029695717560320, journalId=1311385739772133495, articleId=1312006310527332698, 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=1312006315577274784, tenantId=1146029695717560320, journalId=1311385739772133495, articleId=1312006310527332698, authorId=1312006315505971614, language=EN, stringName=Fudong YU, firstName=Fudong, middleName=null, lastName=YU, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=2, address=2.Jilin Sino Agri Sunshine Data Co., Ltd., Changchun 130022, China, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null), CN=AuthorExt(id=1312006315648577953, tenantId=1146029695717560320, journalId=1311385739772133495, articleId=1312006310527332698, authorId=1312006315505971614, language=CN, stringName=于福东, firstName=null, middleName=null, lastName=null, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=2, address=2.吉林省中农阳光数据有限公司,长春 130022, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null)}, companyList=[AuthorCompany(id=1312006314499338630, tenantId=1146029695717560320, journalId=1311385739772133495, articleId=1312006310527332698, xref=2., ext=[AuthorCompanyExt(id=1312006314503532935, tenantId=1146029695717560320, journalId=1311385739772133495, articleId=1312006310527332698, companyId=1312006314499338630, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=2.Jilin Sino Agri Sunshine Data Co., Ltd., Changchun 130022, China), AuthorCompanyExt(id=1312006314511921544, tenantId=1146029695717560320, journalId=1311385739772133495, articleId=1312006310527332698, companyId=1312006314499338630, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=2.吉林省中农阳光数据有限公司,长春 130022)])]), Author(id=1312006315719881123, tenantId=1146029695717560320, journalId=1311385739772133495, articleId=1312006310527332698, 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=1312006315795378597, tenantId=1146029695717560320, journalId=1311385739772133495, articleId=1312006310527332698, authorId=1312006315719881123, language=EN, stringName=Xinhui ZHONG, firstName=Xinhui, middleName=null, lastName=ZHONG, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=2, address=2.Jilin Sino Agri Sunshine Data Co., Ltd., Changchun 130022, China, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null), CN=AuthorExt(id=1312006315891847590, tenantId=1146029695717560320, journalId=1311385739772133495, articleId=1312006310527332698, authorId=1312006315719881123, language=CN, stringName=钟馨慧, firstName=null, middleName=null, lastName=null, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=2, address=2.吉林省中农阳光数据有限公司,长春 130022, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null)}, companyList=[AuthorCompany(id=1312006314499338630, tenantId=1146029695717560320, journalId=1311385739772133495, articleId=1312006310527332698, xref=2., ext=[AuthorCompanyExt(id=1312006314503532935, tenantId=1146029695717560320, journalId=1311385739772133495, articleId=1312006310527332698, companyId=1312006314499338630, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=2.Jilin Sino Agri Sunshine Data Co., Ltd., Changchun 130022, China), AuthorCompanyExt(id=1312006314511921544, tenantId=1146029695717560320, journalId=1311385739772133495, articleId=1312006310527332698, companyId=1312006314499338630, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=2.吉林省中农阳光数据有限公司,长春 130022)])])]
田振江1 , 韩成1 , 张超1, 李月莹1, 于福东2, 钟馨慧2
作者信息
  • 1.长春理工大学计算机科学技术学院,长春 130012
  • 2.吉林省中农阳光数据有限公司,长春 130022
通讯作者:
韩成(1978—),男,教授,主要从事计算机视觉研究,E-mail:
作者简介:

田振江(1998—),男,博士生,主要从事牛脸面部识别研究,E-mail:

Individual Identification Method of Simmental Cattle Based on Improved YOLO v8 Keypoint Detection
Zhenjiang TIAN1 , Cheng HAN1 , Chao ZHANG1, Yueying LI1, Fudong YU2, Xinhui ZHONG2
Affiliations
  • 1.School of Computer Science and Technology, Changchun University of Science and Technology, Changchun 130012, China
  • 2.Jilin Sino Agri Sunshine Data Co., Ltd., Changchun 130022, China
出版时间: 2025-12-25 doi: 10.6041/j.issn.1000-1298.2025.12.054
文章导航
收藏切换

随着畜牧业的智能化发展,基于计算机视觉技术的牛只个体识别在牛场养殖中扮演着越来越重要的角色。然而,在实际应用中,牛脸角度的快速变化、牛只数量的频繁变动以及模型识别速度的局限性严重影响了识别的准确性和效率。为了应对这些挑战,本文提出了一种基于改进YOLO v8关键点检测的西门塔尔牛个体识别方法。该方法包括2个阶段,在牛脸检测阶段,改进YOLO v8模型自动检测牛脸图像中的关键点,并进行矫正对齐和背景裁剪,使牛脸图像角度趋于一致。在牛脸识别阶段,引入了深度可分离卷积以减少参数量,利用残差连接和注意力机制增强了模型的特征提取能力,从而提升了识别准确性和速度。为验证方法的有效性,构建了包含159头西门塔尔牛的牛脸图像数据集。实验结果表明,本文模型在参数量和推理时间上有显著改进,单幅图像处理时间缩短至39 ms以内,识别准确率为95.8%,比其他模型相比平均提高4.8个百分点。此外,本文方法支持实时更新牛只数据库,在增加牛脸图像后,仍能保持较高准确率。

牛脸识别  /  计算机视觉  /  YOLO v8  /  关键点检测

With the intelligent development of animal husbandry, computer-vision-based individual cattle recognition is playing an increasingly important role in modern cattle farming. However, in practical applications, the rapid changes in the angles of cattle faces, frequent fluctuations in the number of cattle, and limitations in model recognition speed severely affect the accuracy and efficiency of recognition. To address these challenges, an individual recognition method for Simmental cattle was proposed based on improved YOLO v8 keypoint detection. The method consisted of two stages: in the cattle face detection stage, the improved YOLO v8 model automatically detected keypoints in cattle face images and performed correction alignment and background cropping to make the angles of the cattle face images more consistent. In the cattle face recognition stage, depthwise separable convolution was firstly introduced to reduce the number of parameters, and then residual connections and attention mechanisms were utilized to enhance the model's feature extraction capabilities, thereby improving recognition accuracy and speed. To validate the effectiveness of the method, a dataset of cattle face images containing 159 Simmental cattle was constructed. Experimental results showed that the proposed model had significant improvements in the number of parameters and inference speed, with processing time for a single image reduced to within 39 ms and a recognition accuracy of 95.8%, an average improvement of 4.8 percentage points compared with that of other models. Additionally, the proposed method supported real-time updates to the cattle database, maintaining high accuracy even after adding new cattle face images.

cow face recognition  /  computer vision  /  YOLO v8  /  key point detection
田振江, 韩成, 张超, 李月莹, 于福东, 钟馨慧. 基于改进YOLO v8关键点检测的西门塔尔牛只个体识别方法. 农业机械学报, 2025 , 56 (12) : 591 -602 . DOI: 10.6041/j.issn.1000-1298.2025.12.054
Zhenjiang TIAN, Cheng HAN, Chao ZHANG, Yueying LI, Fudong YU, Xinhui ZHONG. Individual Identification Method of Simmental Cattle Based on Improved YOLO v8 Keypoint Detection[J]. Transactions of the Chinese Society for Agricultural Machinery, 2025 , 56 (12) : 591 -602 . DOI: 10.6041/j.issn.1000-1298.2025.12.054
畜牧业作为农业的重要组成部分,目前正向规模化、标准化、产业化方向发展,现代化规模养殖逐渐取代个体户散养的养殖方式[1-2]。在现代化肉牛养殖中,通过牛只个体识别技术为每头牛建立数字身份,实现精准的牛只个体追踪与信息管理,发现牛只潜在健康问题,提高养殖场养殖效率。此外,牛只个体识别技术在减少保险纠纷中发挥着关键作用,其可以为每头牛提供准确、唯一的身份标识,能够有效防止骗保和错赔现象,保障保险公司与养殖户的权益。
牛只个体身份识别方法分为接触式和非接触式2种:接触式识别方法包括传统标记方法和电子耳标。传统标记方法虽易于实施,但对牛只健康有影响且易被篡改;电子耳标[3]利用射频识别(Radio frequency identification,RFID)技术,识别速度快、准确性高,但受监测数量、成本和距离等限制,难以大规模推广。非接触式方法如生物特征识别和牛脸识别逐渐成熟,牛只虹膜[4]、视网膜[5]和鼻子纹理[6]等生物特征与人类眼部相似,都是独一无二的,可用于牛只个体身份识别。一些研究[7-10]通过各种策略去定位牛脸虹膜位置和虹膜的特征信息,都取得了较高的识别效果。尽管这些方法准确率高,但采集困难且不易推广。牛脸识别是一种简单、无创、易操作的方法,可以更易区分毛发纹理和面部结构特征。一些研究通过图像处理和神经网络[11]、局部二值模式纹理特征[12]和牛脸关键点[13]方法对牛脸面部特征进行识别,验证了面部特征识别方法的可行性。随着深度学习技术的发展,卷积神经网络在生物识别领域也得到应用和发展。一些研究[14-19]使用LeNet-5、VGG、Vision-Transformer等网络模型进行牛脸面部特征提取,实现牛脸在遮挡和脏污场景下的识别。最近一些研究聚焦多角度牛脸识别[20]和实时视频识别[21],旨在解决由于角度变化导致的牛脸特征信息不完整问题。
虽然基于深度学习技术的牛脸识别在图像识别领域表现出色,但在实际应用中,牛脸角度的快速变化、牛只数量的频繁变动(通常因购买或出售而增加或减少)以及模型识别速度的限制,往往会影响识别效果。随着牛场规模的不断扩大,需要牛脸识别技术能够快速处理这些数量变化。然而,现有的一些深度学习模型在应对牛只增减方面存在明显不足,常依赖于训练过的模型进行概率匹配。当新的牛只加入时,这些方法无法有效区分该牛只是否已存在于数据库中,导致识别的准确性和灵活性受到限制。特别是在新的牛只验证匹配时,常会出现误识别问题。
针对上述问题,本文提出一种基于改进YOLO v8关键点检测的西门塔尔牛只个体识别方法。在牛脸数据采集后,使用牛脸关键点检测模型,实现牛脸图像自动识别,并自动进行矫正对齐裁剪处理,减少背景干扰和角度变化带来的误差。在牛脸识别过程中,构建牛脸识别模型在降低模型参数量的同时,提升处理速度和识别准确性。此外,提出的方法能够识别和处理不在数据库中的牛只,在增加牛脸图像后,本方法仍能保持较高的准确率,提高在实际应用中的灵活性。
数据于2023年7—11月在吉林省德惠牛场拍摄采集,具体拍摄牛场环境如图1所示。数据采集使用普通手机拍摄牛头视频,视频保存格式为MP4,拍摄时长为5~15 s。通过不同距离拍摄数据分析,确定最佳拍摄距离为牛脸正前方约0.7 m,与牛头位置水平,以确保图像清晰度较高,视频拍摄角度以牛只正脸为主。由于拍摄条件限制,数据采集主要在室内进行,采集的牛只品种为西门塔尔牛,共采集159头牛只图像和视频数据。
从采集的牛只中选取85头牛的图像数据进行手工关键点标注,其中训练集包含60头牛的300幅图像,测试集包含25头牛的75幅图像,测试集数据记为CowK。数据标注使用LabelMe软件进行手动标注,标注完成后使用图像平移、增加噪点、遮挡和旋转等数据增强手段,将训练集扩充至1 500幅图像。标注关键点位置如图2所示,图2b标注位置为参照文献[13]中的关键点标注位置,其中1~19为牛脸轮廓关键点,20~23为牛鼻关键点,24~29为左右眼关键点。根据预实验结果,关键点数量会直接影响牛脸裁剪任务的效果,从而影响后续的牛脸识别准确性。所以本文在29个关键点的基础上在牛鼻、左右脸中部分割线等重点区域添加了更多的关键点位置,增加这些区域使模型能够更好地理解牛脸的几何结构,从而提高牛脸裁剪任务的效果,最终标注如图2c所示的43个关键点。其中1~19为牛脸轮廓关键点,20~29为牛鼻关键点,30~33为左眼关键点位置,34~37为右眼关键点位置,38~43为牛只左右脸中部分割线。标注完成后,选择1幅标注好的牛脸正脸图像作为标准蒙版图像,标准蒙版图像包含43个关键点的标注位置,这个蒙版图像和标注位置用于后续的对齐矫正处理。
牛脸关键点的定位可以提供关于牛脸结构和形态的关键信息,因此需要1个能够高效、快速且精确检测牛脸关键点的模型。YOLO v8模型虽在关键点检测任务上表现优异,但由于其模型规模较大,在资源受限设备上运行时可能会受到限制。为了提升关键点模型在这些环境中的实用性,本文对YOLO v8网络模型进行轻量化处理,使其在保持较高识别精度的同时,显著降低计算资源的需求。整体模型如图3所示,主干网络采用MobileNetV3[22]结构,使用轻量级的深度可分离卷积(Depthwise separable convolution,DSConv)和残差块,加速模型收敛,降低模型参数量。同时,在主干网络的最后3层引入卷积块注意力模块(Convolutional block attention module,CBAM)[23]注意力机制,增强特征提取能力和模型表达能力。颈部网络采用2个上采样和特征融合模块组,以捕获更细节的牛脸特征。头部网络连接整个模型的输出,利用主干和颈部网络提取的特征,进行训练并做出预测。
牛脸面部裁剪矫正过程为:①使用关键点检测模型对牛脸图像进行牛脸关键点检测。②将识别后的关键点与标准蒙版图像中的关键点进行对齐操作,确保两者之间位置一致。③对齐后对图像进行仿射变换,将整个牛脸图像按照对齐的比例与蒙版图像进行贴合,使牛脸各部分区域与蒙版图像区域保持一致,实现统一的牛脸形状和位置标准化,减少后续裁剪图像过程中的差异性。
在裁剪牛脸过程中,根据关键点坐标计算牛脸裁剪区域,并从原图像中裁剪出这个区域,对裁剪出的图像进行缩放,使其符合预定的标准尺寸。在缩放后的图像上,通过指定的关键点索引,连接关键点形成面部轮廓,然后提取出轮廓内的牛脸图像,处理完成后获得具有统一视角和标准尺寸的牛脸图像,为后续牛脸识别提供更清晰、更规范的输入。
视频数据采集完成后,使用牛脸关键点检测模型对视频中的牛脸进行识别,自动将每帧识别到的牛脸图像转换为牛脸图像数据,根据检测到关键点进行裁剪对齐矫正,裁剪后尺寸为415像素×295像素,为了确保图像质量,在裁剪过程对牛脸图像数据进行模糊度和相似度筛选,图像的模糊程度使用图像的拉普拉斯边缘强度算法评估,删除平均边缘强度小于2的模糊图像。对于尺寸x×y图像,模糊检测公式为
式中
      BlurScore——图像模糊分数
      x、y——图像宽度、高度像素数量
      i、j——图像行、列坐标位置
      Laplacian(·)——图像拉普拉斯变换值[24]
牛脸相似度使用图像结构相似度(Structural similarity index measure,SSIM)[25]评估,图像m和图像n的SSIM公式为
式中
      μm、μn——图像m、n像素均值
      、——图像m、n像素方差
      σmn——图像m、n协方差
      c1、c2——常数,防止分母为0
为确保同一头牛有更高的区分度,本文对相似度大于0.5的图像进行删除,确保所有图像两者之间的平均相似度在0.5以下。在进行模糊度和相似度筛选后,对于牛脸图像数量不足40幅的牛只,进行删除并不再用于后续处理,确保数据集的质量和模型训练的有效性。
根据关键点检测模型分别制作了以下3个图像数据集:29个关键点裁剪图像数据集,通过使用29个关键点检测模型进行关键点检测,提取牛脸关键点轮廓区域内图像;43个关键点裁剪图像数据集,通过使用43个关键点检测模型进行关键点检测,提取牛脸关键点轮廓区域内图像;无裁剪图像数据集,仅通过43个关键点检测算法对牛脸位置进行识别,保存牛脸识别框内图像。3个数据集图像如图4所示,其中原始图像尺寸为1 280像素×760像素,无裁剪图像尺寸为415像素×415像素,29个关键点和43个关键点裁剪图像尺寸为415像素×295像素。
为了验证本文提出的牛脸识别方法,将数据集划分为原有牛只数据集和新入库牛只数据集,原有牛只数据集共109头牛,每头牛只分别选择80幅图像,按比例14∶3∶3进行划分,其中56幅用于训练牛脸面部特征模型,12幅图像用于将牛只注册到数据库,12幅图像用于验证匹配数据库。对训练集进行数据增强操作,最终训练图像数量共30 520幅,记为CowTr;注册图像数量共1308幅,记为CowTre;测试图像数据共1 308幅,记为CowTc。测试图像与注册图像进行匹配评估模型识别效果。新入库牛只数据集一共50头,每头牛只提取40幅图像,然后从中选择12幅图像与注册的牛只进行匹配。测试图像共600幅,记为CowF,评估牛脸识别模型在验证新牛只图像时的准确性和可靠性。
牛脸识别技术可定义为输入牛脸图像,提取牛脸特征,并与数据库中的已知牛脸特征进行匹配。在第1阶段,本文方法可以自动检测牛脸图像中的关键点位置,进行矫正对齐,裁剪背景区域,从而降低与数据库牛脸图像的差异。在第2阶段,使用FaceNet[26]框架进行牛脸识别任务。输入图像使用第1阶段处理后的牛脸裁剪图像,以特殊的三元组数据导入模式作为牛脸特征提取网络输入,使同一身份的牛脸特征在特征空间中更加接近,而不同身份的牛脸特征更加分散。牛脸图像经特征提取网络处理后,对提取的特征进行L2归一化,最终获得128维牛脸特征向量。最后,使用牛脸特征提取模型对注册集和测试集进行特征编码,完成牛脸数据的注册。然后通过计算测试集编码与注册集编码之间的欧氏距离,进行特征匹配并输出识别结果。整个牛脸识别任务框架如图5所示。
牛脸特征提取网络使用改进的Inception-ResNetv2[27]网络结构,为提高模型识别速度,引入Inception架构、残差网络和DSConv。
将Inception模块的输出添加到输入中,统一输入与输出维度,便于进行残差加法运算。由于裁剪后牛脸图像不包含背景,需要对牛脸的细节特征进行精准捕捉。为此,在Inception-resnet-A、B、C 3个模块后加入CBAM注意力机制,以增强特征提取能力。改进Inception-ResNetv2网络包括6个模块:Stem、Inception-resnet-A、Reduction-A、Inception-resnet-B、Reduction-B和Inception-resnet-C。Stem模块负责对输入进行初步处理和特征提取;Inception-resnet-A、B、C模块分别在不同层次上提取特征;Reduction-A和Reduction-B模块用于缩减特征图尺寸,从而减少计算量和参数量。这种设计可有效减少模型参数量,加速模型训练和推理,同时在特征提取能力和模型识别性能上有显著提升。网络结构如图6所示。
在牛脸识别方法中,损失函数由三元组损失和交叉熵损失共同组成。通过结合这两种损失,模型不仅能够有效地学习样本间的相对距离,还能进行精确的分类,三元组损失[28]是一种用于训练人脸识别和其他识别学习任务的常用损失函数。该函数基于三元组样本,对于牛脸数据集中,在每个样本设置锚点牛脸图像、正例牛脸图像和负例牛脸图像,正例图像与锚点图像同一身份,而负例图像与锚点图像不同身份。整体损失函数的目标是使得锚点图像与其正例图像之间的距离缩小,与负例图像之间的距离增加,三元组损失定义为
式中
      o——训练样本数量
      ———第k个样本锚点图像
      、——第k个样本正例、负例图像
      ——欧几里得距离
      h(·)——提取特征的函数
      α——间隔的超参数,规定正图像与负图像之间的最小距离
      [·]+——取正值函数,即当括号内值为负数时取0,否则取该值本身
这样的训练目标有助于网络能够将同一牛脸图像之间的特征向量尽可能靠近,不同牛脸图像之间的特征向量尽可能远离,从而实现更好的牛脸验证性能,三元组损失示意图如图7所示。
交叉熵损失用于分类任务,通过最小化预测概率分布与真实分布之间的交叉熵来训练模型。模型在输出层使用Softmax函数,计算每个类概率,并通过负对数似然损失来计算损失值。
实验环境:CPU配置为13th Gen Intel(R)Core(TM)i9-13900HX,GPU为GeForce RTX4080,具备12 GB显存和32 GB内存,操作系统为Windows 11系统。采用Pytorch深度学习框架,对模型训练学习率动态调整设值,范围为0.001~0.000 01,批量大小为12,迭代次数为100。采用自适应矩阵估计算法对模型进行优化,损失值计算则采用三元组损失函数和交叉熵损失结合方法。
在牛脸关键点检测阶段,使用归一化均值误差(Normalized mean error,NME)算法计算牛脸关键点坐标与真实牛脸坐标之间的误差,衡量关键点检测算法准确度。
使用牛脸左右眼距(眼中心距离)作为归一化因子。在牛脸识别阶段时采取准确率(Accuracy)、精确率(Precision)、召回率(Recall)、假正率(FPR)和F1分数(F1 score)对本文模型进行对比验证。
为了全面评估模型的推理性能,本文统计模型的参数量和推理时间。其中,推理时间指模型在单次推理时消耗的时间,而平均推理时间则为多次推理后平均消耗时间,用于衡量模型实时性表现。
使用目标关键点相似度(Object keypoint similarity,OKS)作为关键点检测精度的评价指标[29]。OKS能够衡量检测到的关键点与真实关键点之间的相似度,并基于不同阈值计算精确率、召回率和平均精度,从而全面评估模型在关键点检测任务中的表现。
在实验部分除3.3节使用29个关键点裁剪图像数据集和无裁剪图像数据集外,其余实验部分均使用43个关键点裁剪图像数据集。
在牛脸关键点检测过程中,采用NME算法评估模型性能,其中归一化因子d选取牛脸左右眼4个关键点中心点之间的距离,避免不同牛脸大小和姿态对结果的干扰,确保评估一致性和可靠性。对比YOLO v8模型与本文模型在牛脸正面关键点识别任务中的表现,如表1所示。
为了验证NME指标对后续牛脸识别精度的影响,在相同的43个关键点基础下,对基于YOLO v8模型的裁剪数据和基于本文模型的裁剪数据分别进行了实验。实验结果表明,在牛脸关键点检测任务中,YOLO v8与本模型NME差值0.41个百分点,对应的识别准确率相差约0.1个百分点。但是本文方法推理时间10 ms与改进前相比提升60%,参数量下降51.6%。因此,虽然原模型在精度上略有优势,但改进模型在计算效率和资源占用上显著优于原模型,适合实际应用场景中对实时性和设备资源要求较低的任务。
为了全面评估本文提出的关键点裁剪矫正技术对牛脸识别结果的影响,分别对29个关键点裁剪图像数据集、43个关键点裁剪图像数据集以及无裁剪图像数据集进行了训练实验。在实验评估过程,除数据集类型不同外,其余实验条件均保持一致,数据划分比例也与43个关键点裁剪数据集的划分比例一致,以确保结果的公平性与可比性。通过不同裁剪方案的对比分析,可以更准确地了解关键点数量和裁剪策略对模型性能的影响,结果见表2。
由表2可知,未使用裁剪技术的模型在处理牛脸图像时,受图像背景和其他干扰因素的影响较大,各指标均最低。使用29个关键点进行裁剪后,模型能够更好地聚焦于牛脸的关键特征,减少了背景干扰,显著提高了整体各指标。使用43个关键点进行裁剪后,模型能够更加精细地提取牛脸边缘的特征信息,模型的表现进一步提升,达到最佳效果。综上所述,裁剪矫正技术在牛只面部个体识别模型的准确性和稳健性具有积极作用,而且关键点数量也会影响最终的识别效果。
在图像经过43个关键点模型矫正裁剪处理后,所有牛脸的姿态和大小均与标准牛脸相似,整体特征相似性较高。本文选择匹配欧氏距离阈值为0.6进行特征提取和验证匹配,在相同实验环境和参数设置下,将FaceNet结构中主干网络分别替换为MobileNetv2[30]、Inception-ResNetv2[27]、VGG1[31]、Vision-Transformer[32]、Swin-Transformer[33]和ResNet50[34]网络进行对比,不同网络模型对比结果如表3所示。由表3可知,本文模型较其他模型的牛脸识别准确率、召回率和F1分数均更优,本文模型的精确率和假正率没有取得最优结果,经过分析发现Swin-Transfomer模型的精确率和假正率都高于本模型,尽管Swin-Transformer在指标上表现较好,但其整体准确率、召回率、F1分数仍不及本文模型,且该模型的推理时间较慢,不适合于实时牛脸识别任务,本文模型在识别牛脸时更加全面稳定。
在本实验中,对数据库中添加不同规模的牛只数据对模型准确率的影响进行评估,如表4所示,即在数据库中分别添加50头牛的50幅图像、100幅图像、250幅图像、350幅图像和600幅图像,以进行对比验证。测试数据集使用数据库中的牛只验证集CowTc,与之进行匹配数据使用数据库中的牛只注册集CowTre和新添加的牛只图像。通过不同规模批次的实验,分析数据规模对各网络模型识别准确率的影响。
由表4可知,不同模型在增加图像数量后的表现存在显著差异,MobileNetv2、VGG16、ResNet50模型在增加图像数量后,准确率基本保持不变,可能是由于这些模型对数据集变化具有较高的鲁棒性,能够在不同数量图像上保持稳定性能,但这些模型的准确率相对较低。而Inception-ResNetv2、Vision-Transformer、Swin-Transformer模型和本文模型,在增加图像数量后,模型对数据集变化较为敏感,准确率有所下降。本文模型从96.4%下降到95.3%,始终高于其他模型准确率。这表明模型在处理一定规模的数据变化时仍保持了较高准确率,但也反映了模型在应对更多样化数据时可能面临一定挑战,如在实际应用中加入大规模数据,可能需要对新加入的牛脸图像进行训练处理以提升牛脸识别模型准确率。
为验证本文模型引入的DSConv和CBAM注意力机制模块对牛脸识别模型的影响,本文在与3.3节相同的实验环境和参数设置下进行消融实验,结果如表5所示。
相较于Inception-ResNetv2(基础模型),将DSConv引入到模型中,模型参数量大幅减少,同时识别速度提升,表明DSConv的引入有效降低了模型计算复杂度,提高了模型运行效率,但模型精确率和假正率有所下降,反映出在降低计算成本的同时,模型的识别精度也受到了一定影响。通过将CBAM注意力机制加入到Inception-ResNetv2+DSConv模型后,模型整体性能得到显著提升。所有指标均达到最高值,虽然模型参数量有所上升,但仍优于Inception-ResNetv2模型,总体表明引入DSConv和CBAM注意力机制适合应用于对识别精度要求较高的场景。
整个牛脸识别过程推理时间和参数量如表6所示。由表6可知,本文模型可显著减少总推理时间,与任意牛脸特征模型组合,总推理时间均明显低于使用YOLO v8的组合。虽然MobileNetv2模型推理时间低于本文模型,但是模型的准确率明显低于本文模型。尽管Vision-Transformer和Swin-Transformer在牛脸识别任务中有较高的识别精度,但其总推理时间较长,不适用于实时性要求较高的应用场景。综合以上评估结果,本文模型在牛脸识别任务中,整体表现更优。
牛脸图像易受图像质量、栏杆遮挡和牛脸姿态等外界因素影响,导致关键点检测不准确或缺失,从而影响姿态对齐的效果。本文模型在采集牛脸视频数据时,会对拍摄位置和角度提出限制,影响获取较高质量的牛脸图像。
采用改进YOLO v8关键点检测模型,在关键点检测模型训练中采用局部遮挡数据增强方法,增强模型在遮挡情况下检测鲁棒性。此外,在检测过程如果牛脸关键特征足够清晰,即使存在部分遮挡,该模型仍能准确识别所有关键点。然而,当牛脸姿态偏离正面过大或牛脸特征被大范围遮挡时,模型的关键点检测会整体失败,而非仅限于部分关键点的缺失。因此在处理牛脸视频数据时,关键点检测模型会自动删除无法识别的牛脸图像。这种筛选机制有助于提高后续姿态对齐的质量,但也可能导致可用的训练数据量减少,影响模型的训练效果和泛化能力。未来研究中,将进一步探索多角度数据采集及处理方法,优化模型算法,力求在去除低质量数据的同时,最大限度提高在不同姿态和遮挡情况下的检测精度,保留更多有用的训练样本。
为了验证关键点检测模型的准确性,在数据采集中采集部分受遮挡的牛脸图像,通过分析这些遮挡数据特点,使用牛脸关键点测试集图像,构建牛脸遮挡数据集,如图8所示。该数据集模拟不同遮挡位置、形状和范围,遮挡比例为5%~50%。
在不同遮挡条件下对模型关键点检测平均精度进行评估,结果如表7所示。随着遮挡比例增加,模型关键点检测精度逐渐下降。特别是在牛只面部内部单个遮挡和多个遮挡情况下,平均精度在遮挡区域不断扩大情况下逐渐降低。在50%遮挡时降低至50%以下,且有超过1/3的牛只在该条件下无法正确识别。相比之下,当遮挡发生在牛只面部外部区域时,模型表现相对稳定,平均精度在不同遮挡比例下均保持较高水平,说明牛脸面部区域外的遮挡对模型检测影响较小。
结合实验结果可知,大面积遮挡会导致关键点不准确和无法识别的问题,影响牛脸姿态对齐效果。因此,未来研究将重点关注对遮挡区域和缺损区域的处理,进一步提高模型在复杂环境下的检测准确性和姿态对齐效果。
在牛只个体识别研究中,本文模型对拍摄位置提出了限制条件。在此条件下采集牛脸图像具有较高的质量和显著的特征,有助于提升关键点检测的准确性。通过精确裁剪、对齐和矫正,减少图像间差异和背景干扰。结合FaceNet进行特征提取和识别,在西门塔尔牛品种个体识别中,本文模型识别准确率达到95.8%,且单幅图像处理时间缩短至39 ms以内,满足快速处理需求。
为进一步验证本文模型在不同场景和不同牛种中的适用性,对多种场景与不同品种牛只进行实验,结果如图9所示。
从图9可知,本文模型在牛只面部图像质量较高的情况下具备较好的识别表现。然而,由于本文所使用数据集牛只品种为西门塔尔牛,且训练数据主要集中于正脸角度,因此在面对牛种差异较大或面部角度显著偏离正面的情况下,本文模型无法实现牛脸面部位置定位和关键点检测。此外,本文模型在复杂场景下仍存在一定局限性,比如,当牛只面部被大面积遮挡或光照条件极端时(过亮或过暗),关键点检测的准确性会大幅下降,导致姿态对齐效果不佳,进而影响最终的识别结果。
(1)提出了基于改进YOLO v8关键点检测的西门塔尔牛只个体识别方法,通过移动手机设备采集真实牛场环境视频图像,利用改进YOLO v8模型关键点检测技术自动识别牛脸关键点位置,对关键点区域进行矫正对齐和背景裁剪处理,对处理后的数据进行数据增强操作,制作了牛脸图像数据集。
(2)提出的关键点检测模型NME值为3.52%,误差较低。与原YOLO v8模型相比虽有下降,但对牛脸识别影响不大,且改进模型的参数量和推理时间都有明显提升。本文提出的牛脸识别模型结合残差网络、深度可分离卷积和注意力机制,实现了对牛只个体身份的准确识别。与其他同类模型相比,本文模型识别准确率达到95.8%,证明了在牛脸识别任务中的有效性。
(3)牛脸识别性能实验结果表明,在牛脸图像数量变化的情况下,仍有较高的准确率,并可使单幅图像处理时间缩短到39 ms以内,有助于模型在移动端等计算量小的设备上应用。
  • 吉林省重点科技研发项目(20180201069GX)
参考文献 引证文献
排序方式:
[1]
张姝瑾, 许兴时, 邓洪兴, 等. 基于YOLO v8n-seg-FCA-BiFPN的奶牛身体分割方法[J]. 农业机械学报, 2024, 55(3): 282-289, 391.
ZHANG Shujin, XU Xingshi, DENG Hongxing, et al. Segmentation model of cow body parts based on YOLO v8n-seg-FCA-BiFPN[J]. Transactions of the Chinese Society for Agricultural Machinery, 2024, 55(3): 282-289, 391. (in Chinese)
[2]
GUO Y, HONG W, WU J, et al. Vision-based cow tracking and feeding monitoring for autonomous livestock farming: the YOLOv5s-CA+DeepSORT-Vision Transformer[J]. IEEE Robotics & Automation Magazine, 2023, 30(4): 68-76.
[3]
董正奇. 基于体温监测和饮水行为记录的牛只监管耳标设计[D]. 杨凌: 西北农林科技大学, 2024.
DONG Zhengqi. Cow regulatory ear tag design based on body temperature monitoring and drinking behavior records[D]. Yangling: Northwest A&F University, 2024. (in Chinese)
[4]
司永胜, 王朝阳, 张艳, 等. 基于深度与传统特征融合的非限制条件下奶牛个体识别[J]. 农业机械学报, 2023, 54(6): 272-279.
SI Yongsheng, WANG Zhaoyang, ZHANG Yan, et al. Individual identification of dairy cows under unrestricted conditions based on fusion of deep and traditional features[J]. Transactions of the Chinese Society for Agricultural Machinery, 2023, 54(6): 272-279. (in Chinese)
[5]
LOPEZ R B. Diseases of the heart and blood vessels and their relation to the eye[J]. Clinical and Experimental Optometry, 1943, 26(2): 60-69.
[6]
KUMAR S, SINGH S K, ABIDI A I, et al. Group sparse representation approach for recognition of cattle on muzzle point images[J]. International Journal of Parallel Programming, 2018, 46(5): 812-837.
[7]
ALLEN A, GOLDEN B, TAYLOR M, et al. Evaluation of retinal imaging technology for the biometric identification of bovine animals in Northern Ireland[J]. Livestock Science, 2008, 116(1-3): 42-52.
[8]
孔强, 赵林度. 虹膜识别在肉类食品安全追溯系统中的应用及关键技术研究[J]. 中国安全科学学报, 2009, 19(3): 155-160.
KONG Qiang, ZHAO Lindu. Research on the application of iris recognition in meat-food supply chain management[J]. China Safety Science Journal, 2009, 19(3): 155-160. (in Chinese)
[9]
盛大玮. 牛眼虹膜识别技术研究[D]. 上海: 华东师范大学, 2010.
SHENG Dawei. Research on technology of cattle's iris recognition[D]. Shanghai: East China Normal University, 2010. (in Chinese)
[10]
李超, 赵林度. 牛眼虹膜定位算法研究及其在肉食品追溯系统中的应用[J]. 中国安全科学学报, 2011, 21(3): 124-130.
LI Chao, ZHAO Lindu. Research on cattle iris localization algorithm and its application in meat food tracking and traceability system[J]. China Safety Science Journal, 2011, 21(3): 124-130. (in Chinese)
[11]
KIM H T, IKEDA Y, CHOI H L. The identification of Japanese black cattle by their faces[J]. Asian-Australasian Journal of Animal Sciences, 2005, 18(6): 868-872.
[12]
CAI C, LI J. Cattle face recognition using local binary pattern descriptor[C]//2013 Asia-Pacific Signal and Information Processing Association Annual Summit and Conference. IEEE, 2013: 1-4.
[13]
宋肖肖. 牛脸特征点检测的研究与实现[D]. 杨凌: 西北农林科技大学, 2017.
SONG Xiaoxiao. Research and implementation of cattle face feature points detection[D]. Yangling: Northwest A&F University, 2017. (in Chinese)
[14]
孙曦, 赵平. 基于卷积神经网络的牛脸识别方法研究[J]. 大众标准化, 2020(5): 59-60.
[15]
黄俊华, 田壮. 基于改进VGG卷积神经网络的奶牛牛脸识别研究[J]. 湖北农机化, 2019(13): 126.
[16]
郑鹏, 沈雷, 刘浩, 等. 基于Vision Transformer模型的牛脸识别算法[J]. 杭州电子科技大学学报(自然科学版), 2022, 42(6): 40-46.
[17]
刘世锋, 常蕊, 李斌, 等. 基于脸部RGB-D图像的牛只个体识别方法[J]. 农业机械学报, 2023, 54(增刊1): 260-266
LIU Shifeng, CHANG Rui, LI Bin, et al. Individual identification of cattle based on RGB-D images[J]. Transactions of the Chinese Society for Agricultural Machinery, 2023, 54(Supp. 1): 260-266. (in Chinese)
[18]
杨胜楠, 赵建敏, 杨梅, 等. 基于去偏置项SoftMax和紧致度量损失函数的牛脸识别方法[J]. 黑龙江畜牧兽医, 2024(4): 36-42.
YANG Shengnan, ZHAO Jianmin, YANG Mei, et al. Cattle face recognition method based on debiased term SoftMax and compact metric loss function[J]. Heilongjiang Animal Science and Veterinary Medicine, 2024(4): 36-42. (in Chinese)
[19]
叶孟珂, 李宝山, 杨梅, 等. 基于深度学习的牛脸识别系统设计[J]. 黑龙江畜牧兽医, 2024(4): 43-48.
YE Mengke, LI Baoshan, YANG Mei, et al. Design of cattle face recognition system based on deep learning[J]. Heilongjiang Animal Science and Veterinary Medicine, 2024(4): 43-48. (in Chinese)
[20]
WENG Z, MENG F, LIU S, et al. Cattle face recognition based on a two-branch convolutional neural network[J]. Computers and Electronics in Agriculture, 2022, 196: 106871.
[21]
BERGMA N, YITZHAKY Y, HALACHMI I. Biometric identification of dairy cows via real-time facial recognition[J]. Animal, 2024, 18(3): 101079.
[22]
HOWARD A, SANDLER M, CHU G, et al. Searching for MobileNetv3[C]//Proceedings of the IEEE/CVF International Conference on Computer Vision, 2019: 1314-1324.
[23]
WOOS, PAPK J, LEE J Y, et al. Cbam: convolutional block attention module[C]//Proceedings of European Conference on Computer Vision, 2018: 3-19.
[24]
PRETUZ S, PUIG D, GARCIA M A. Analysis of focus measure operators for shape-from-focus[J]. Pattern Recognition, 2013, 46(5): 1415-1432.
[25]
WANG Z, BOVIK A C, SHEIKH H R, et al. Image quality assessment: from error visibility to structural similarity[J]. IEEE Transactions on Image Processing, 2004, 13(4): 600-612.
[26]
SCHROFF F, KALENICHENKO D, PHILBIN J. Facenet: a unified embedding for face recognition and clustering[C]//Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, 2015: 815-823.
[27]
SZEGEDY C, IOFFE S, VANHOUCKE V, et al. Inception-v4, inception-resnet and the impact of residual connections on learning[C]//Proceedings of the AAAI Conference on Artificial Intelligence, 2017, 31(1): 4278-4284.
[28]
BREDIN H. TristouNet: triplet loss for speaker turn embedding[C]//Proceedings of the IEEE International Conference on Acoustics, Speech, and Signal Processing (ICASSP). IEEE, 2017: 5430-5434
[29]
宋怀波, 华志新, 马宝玲, 等. 基于SimCC-ShuffleNetV2的轻量化奶牛关键点检测方法[J]. 农业机械学报, 2023, 54(10): 275-281, 363.
SONG Huaibo, HUA Zhixin, MA Baoling, et al. Lightweight keypoint detection method of dairy cow based on SimCC-ShuffleNetV2[J]. Transactions of the Chinese Society for Agricultural Machinery, 2023, 54(10): 275-281, 363. (in Chinese)
[30]
SANDLER M, HOWARD A, ZHU M, et al. Mobilenetv2: inverted residuals and linear bottlenecks[C]//Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, 2018: 4510-4520.
[31]
LIU S, DENG W. Very deep convolutional neuralnetwork based image classification using small training sample size[C]//2015 3rd IAPR Asian Conference on Pattern Recognition (ACPR). IEEE, 2015: 730-734.
[32]
HAN K, WANG Y, CHEN H, et al. A survey on vision transformer[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2022, 45(1): 87-110.
[33]
LIU Z, LIN Y, CAO Y, et al. Swin transformer: hierarchical vision transformer using shifted windows[C]//Proceedings of the IEEE/CVF International Conference on Computer Vision, 2021: 10012-10022.
[34]
HE K, ZHANG X, REN S, et al. Deep residual learning for image recognition[C]//Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, 2016: 770-778.
2025年第56卷第12期
PDF下载
152
69
引用本文
BibTeX
文章信息
doi: 10.6041/j.issn.1000-1298.2025.12.054
  • 接收时间:2024-08-12
  • 首发时间:2026-09-30
  • 出版时间:2025-12-25
补充材料
相关文章
文章信息
作者
出版历史
  • 收稿日期:2024-08-12
  • 修回日期:2024-10-18
基金
吉林省重点科技研发项目(20180201069GX)
作者信息
    1.长春理工大学计算机科学技术学院,长春 130012
    2.吉林省中农阳光数据有限公司,长春 130022

通讯作者:

韩成(1978—),男,教授,主要从事计算机视觉研究,E-mail:
参考文献
分享链接
https://castjournals.cast.org.cn/joweb/nyjxxb/CN/10.6041/j.issn.1000-1298.2025.12.054
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
科
Family
属数
Number of
genus
种数
Number of
species
占总种数比例
Percentage of
total species (%)
属
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