智慧农业（中英文）,2022,4(2):1-18doi:10.12133/j.smartag.SA202204005
专题——智慧畜牧关键技术与装备
2.中国农业大学农业农村部农业信息获取技术重点实验室，北京100083
3.浙大宁波理工学院计算机与数据工程学院，浙江宁波315200
2.KeyLaboratoryofAgriculturalInformationAcquisitionTechnology,MinistryofAgricultureandRuralAffairs,ChinaAgriculturalUniversity,Beijing100083,China
3.SchoolofComputingandDataEngineering,NingboTechUniversity,Ningbo315200,China
收稿日期:2022-04-28
Received:2022-04-28
作者简介Aboutauthors
关键词：奶牛养殖;计算机视觉;生理参数监测;疾病诊断;精细畜牧业;智能养殖
Keywords：dairyfarming;computervisiontechnology;physiologicalparametersmonitoring;diseasesdiagnosing;precisionlivestockfarming;intelligentfarming
本文引用格式
康熙,刘刚,初梦苑,李前,王彦超.基于计算机视觉的奶牛生理参数监测与疾病诊断研究进展及挑战[J].智慧农业（中英文）,2022,4(2):1-18.doi:10.12133/j.smartag.SA202204005
KANGXi,LIUGang,CHUMengyuan,LIQian,WANGYanchao.AdvancesandChallengesinPhysiologicalParametersMonitoringandDiseasesDiagnosingofDairyCowsBasedonComputerVision[J].SmartAgriculture,2022,4(2):1-18.doi:10.12133/j.smartag.SA202204005
图1奶牛生理参数监测技术示意图
Fig.1Sketchmapofmonitoringtechniqueforcowphysiologicalparameters
Table1Researchesoncomputervision-basedmeasurementofbodysizeofdairycows
肩高、臀宽、体宽、坐骨
宽度等
为0.75
综上所述，利用消费级深度相机采集奶牛深度信息，并转换为三维点云数据，通过计算机视觉和深度学习等先进技术，自动计算奶牛体尺参数，成为目前奶牛体尺自动测量领域的主要方法与研究热点。但如何提高点云处理速度、体尺测点定位精度、算法普适性与鲁棒性仍是目前研究亟需突破的关键问题。
图2热红外相机工作原理示意图
Fig.2Schematicdiagramofworkingprincipleofthermalinfraredcamera
Table2Researchesoncomputervision-basedtemperaturedetectionofdairycows
总而言之，基于热红外视频的奶牛体温监测技术相较于人工检测方法具有节省人力物力、无损伤和无应激等优点，是智慧畜牧业中畜禽信息智能感知技术的重要组成部分，但该技术对奶牛体温获取的精度仍有待进一步提高，热红外图像中的目标精准检测以及监测系统对环境鲁棒性的提升是基于热红外视频的奶牛体温监测技术普及应用的关键。
Table3Researchesoncomputervision-basedweightmeasurementofdairycows
Table4Researchesoncomputervision-basedmastitisdetectionofdairycows
Fotric-625c红外
热像仪
奶牛乳房温度拟合线
的斜率
奶牛左右患病乳房识别准确率
为75%
平均相对误差为3.67%，标准差
为1.88%
综上所述，计算机视觉技术可应用于奶牛乳房炎检测工作中，通过对奶牛关键部位的准确定位，提取其温差信息并与所设定的阈值进行比较，实现奶牛乳房炎检测。然而目前基于温度阈值的乳房炎检测方法准确性仍不高，并且易受环境等因素干扰。因此，乳房炎检测方法的研究以及提高系统的鲁棒性是基于计算机视觉的奶牛乳房炎检测技术的关键研究方向。
Table5Researchesonlamenessdetectionofdairycowbasedoncomputervision
灵敏度为100.00%，特异性为75.00%，准确率
为95.70%
注：①k最邻近分类算法（K-NearestNeighbor，KNN）；②双向长短词记忆模型（Bi-directionalLongShort-TermMemory，BiLSTM）；
③支持向量机（SupportVectorMachine，SVM）
图3奶牛跛行检测系统结构示意图
Fig.3Systemschematicofcowlamenessdetection
总而言之，三种相机均可以用于奶牛跛行检测研究，依托不同类型相机所构建的检测系统，具有不同的拍摄角度、数据采集格式和检测方法，并且各有优缺点。不同的研究依据其检测的环境和需求，需要选择不同类型的相机实现跛行检测。
综上所述，基于计算机视觉的奶牛跛行识别已经取得了较好的研究进展，识别准确率以及自动化程度已经显著提高。基于可见光相机、深度相机以及热红外相机的方法都可以较好地实现奶牛跛行检测，但三者各有优势和劣势，适用场景也各不相同，因此在养殖场进行实际应用阶段，需要根据实际情况选取最适合的方案。
（4）检测系统成本较高。检测系统的成本和检测能力是影响其在养殖场使用的两个主要因素。计算机视觉系统主要由相机和计算机（工控机）组成。相机（可见光相机、深度相机和热红外相机）价格跨度较大，尤其是热红外相机目前价格相对较高。但廉价相机采集的数据会为后续的数据处理过程带来困难（如鬼影图像），从而影响检测结果。因此对于检测系统的开发，不仅要保证其检测性能，同时也要充分考虑其价格成本的受众情况。
计算机视觉技术作为中国畜禽养殖业智能化发展的重要环节，可以在奶牛生理参数监测与疾病诊断领域充分发挥其无应激、高效率等优势，为了更好地推动基于计算机视觉的奶牛生理参数监测与疾病诊断研究发展，尽早实现产业化集成和应用推广，本文提出以下发展建议。
王艳阳,李彤,刘佳丽,等.中国奶牛产业发展现状与对策
陈艳飞.奶牛产业发展现状与未来
农业农村部:"十四五"奶业竞争力提升行动方案
李胜利,姚琨,曹志军,等.2021年奶牛产业技术发展报告
刘一明.实施九大任务提振奶业发展——"十四五"奶业竞争力如何提升有了行动方案
刘继芳,韩书庆,齐秀丽.中国信息化畜禽养殖技术应用现状与展望
LIUJ,HANS,QIX.ApplicationprogressandprospectsofthelivestockandpoultryinformatizedbreedingtechnologyinChina
王国占,侯方安,车宇.国内外无人化农业发展状况
高学杰.高产奶牛养殖技术要点与疾病防控
滕光辉.畜禽设施精细养殖中信息感知与环境调控综述
TENGG.Informationsensingandenvironmentcontrolofprecisionfacilitylivestockandpoultryfarming
赵春江.智慧农业发展现状及战略目标研究
ZHAOC.State-of-the-artandrecommendeddevelopmentalstrategicobjectivesofsmartagriculture
韩书庆,张晶,程国栋,等.奶牛跛行自动识别技术研究现状与挑战
HANS,ZHANGJ,CHENGG,etal.Currentstateandchallengesofautomaticlamenessdetectionindairycattle
徐文娇,董彦君,董玉兰,等.不同年龄段奶牛临床参数和血常规生理指标调查分析
XUW,DONGY,DONGY,etal.Differentageparagraphthecowclinicalparametersandroutinebloodphysiologicalindexanalysis
吴宇峰,李一鸣,赵远洋,等.基于计算机视觉的奶牛体况评分研究综述
WUY,LIY,ZHAOY,etal.Reviewofresearchonbodyconditionscorefordairycowsbasedoncomputervision
张小栓,张梦杰,王磊,等.畜牧养殖穿戴式信息监测技术研究现状与发展分析
ZHANGX,ZHANGM,WANGL,etal.Researchstatusanddevelopmentanalysisofwearableinformationmonitoringtechnologyinanimalhusbandry
LIS,YAOK,CAOZ,etal.Technicaldevelopmentreportofthedairyindustryin2021
STAJNKOD,BRUSM,HOEVARM.Estimationofbullliveweightthroughthermographicallymeasuredbodydimensions
TADEMIR,üRKMEZA,NAL.Afuzzyrule-basedsystemforpredictingtheliveweightofHolsteincowswhosebodydimensionsweredeterminedbyimageanalysis
郭浩,张胜利,马钦,等.基于点云采集设备的奶牛体尺指标测量
GUOH,ZHANGS,MAQ,etal.CowbodymeasurementbasedonXtion
MARINELLOF,PEZZUOLOA,CILLISD,etal.ApplicationofKinect-Sensorforthree-dimensionalbodymeasurementsofcows
NIRO,PARMETY,WERNERD,etal.3DComputer-visionsystemforautomaticallyestimatingheiferheightandbodymass
PEZZUOLOA,GUARINOM,SARTORIL,etal.Afeasibilitystudyontheuseofastructuredlightdepth-cameraforthree-dimensionalbodymeasurementsofdairycowsinfree-stallbarns
LECOZLERY,ALLAINC,CAILLOTA,etal.High-precisionscanningsystemforcomplete3Dcowbodyshapeimagingandanalysisofmorphologicaltraits
李琦,刘伟,赵建敏.基于双目视觉及MaskRCNN的牛体尺无接触测量
LIQ,LIUW,ZHAOJ.Non-contactmeasurementofbovinebodysizebasedonbinocularvisionandMaskRCNN
初梦苑.基于三维重建的奶牛体尺检测与体重预估研究
CHUM.Researchonbodymeasurementandweightestimationofcowsbasedon3Dreconstruction
KAMCHENSG,SANTOSEFDOS,LOPESLB,etal.ApplicationofdepthsensortoestimatebodymassandmorphometricassessmentinNelloreheifers
ZHANGJ,ZHUANGY,JIH,etal.Pigweightandbodysizeestimationusingamultipleoutputregressionconvolutionalneuralnetwork:Afastandfullyautomaticmethod
GUOH,MAX,MAQ,etal.LSSA_CAU:Aninteractive3Dpointcloudsanalysissoftwareforbodymeasurementoflivestockwithsimilarformsofcowsorpigs
FORBESA,DEOLIVEIRAM,DENNISMR.Structuredlight
HEINRICHSAJ,ROGERSGW,COOPERJB.PredictingbodyweightandwitherheightinHolsteinheifersusingbodymeasurements
LESOSKYM,DUMASS,CONRADIEI,etal.Aliveweight-heartgirthrelationshipforaccuratedosingofeastAfricanshorthornzebucattle
刘忠超,范伟强,何东健.奶牛体温检测研究进展
LIUZ,FANW,HED.Researchprogressofbodytemperaturedetectionindairycows
何东健,宋子琪.基于热红外成像与骨架树模型的奶牛眼温自动检测
HED,SONGZ.Automaticdetectionofdairycow'seyetemperaturebasedonthermalinfraredimagingtechnologyandskeletontreemodel
张磊,董茹月,侯宇,等.奶牛体温评价指标及测定方法研究进展
ZHANGL,DONGR,HOUY,etal.Researchprogressonevaluationindicesandmeasurementsofbodytemperatureindairycows
CUTHBERTSONH,TARRG,GONZáLEZLA.Methodologyfordataprocessingandanalysistechniquesofinfraredvideothermographyusedtomeasurecattletemperatureinrealtime
WILLARDS,DRAYS,FARRARR,etal.Useofinfraredthermalimagingtoquantifydynamicchangesinbodytemperaturefollowinglipopolysaccharide(LPS)administrationindairycattle
HOFFMANNG,SCHMIDTM,AMMONC,etal.Monitoringthebodytemperatureofcowsandcalvesusingvideorecordingsfromaninfraredthermographycamera
SALLESMSV,SILVASCDA,SALLESFA,etal.Mappingthebodysurfacetemperatureofcattlebyinfraredthermography
PENGD,CHENS,LIG,etal.Infraredthermographymeasuredbodysurfacetemperatureanditsrelationshipwithrectaltemperatureindairycowsunderdifferenttemperature-humidityindexes
何金成,张鲜,李素青,等.环境温湿度及测量部位对奶牛红外热成像温度的影响
HEJ,ZHANGX,LIS,etal.Effectsofambienttemperatureandrelativehumidityandmeasurementsiteonthecow'sbodytemperaturemeasuredbyinfraredthermography
JADDOAMA,GONZALEZL,CUTHBERTSONH,etal.Multivieweyelocalisationtomeasurecattlebodytemperaturebasedonautomatedthermalimageprocessingandcomputervision
张旭东,康熙,马丽,等.基于热红外图像的奶牛乳房炎自动检测方法
ZHANGX,KANGX,MAL,etal.Automaticdetectionmethodofdairycowmastitisbasedonthermalinfraredimage
ZHANGX,KANGX,FENGN,etal.Automaticrecognitionofdairycowmastitisfromerthmalimagesbyadeeplearningdetector
孙雨坤,岳奎忠,李文茜,等.图像信息技术在奶牛生产中的应用
SUNY,YUEK,LIW,etal.Applicationofimageinformationtechnologyindairycowproduction
KUZUHARAY,KAWAMURAK,YOSHITOSHIR,etal.ApreliminarilystudyforpredictingbodyweightandmilkpropertiesinlactatingHolsteincowsusingathree-dimensionalcamerasystem
HANSENMF,SMITHML,SMITHLN,etal.Automatedmonitoringofdairycowbodycondition,mobilityandweightusingasingle3Dvideocapturedevice
TASDEMIRS,URKMEZA,INALS.DeterminationofbodymeasurementsontheHolsteincowsusingdigitalimageanalysisandestimationofliveweightwithregressionanalysis
张立倩.基于模糊逼近计算视觉算法的奶牛体重测量
ZHANGL.Thecowsweightcalculationbasedonnon-measurementmachinevision
SONGX,BOKKERSEAM,VANDERTOLPPJ,etal.Automatedbodyweightpredictionofdairycowsusing3-dimensionalvision
牛金玉.基于三维点云的奶牛体尺测量与体重预测方法研究
NIUJ.Bodysizemeasurementandweightpredictionfordairycowsbasedon3Dpointcloud
LECOZLERY,ALLAINC,XAVIERC,etal.VolumeandsurfaceareaofHolsteindairycowscalculatedfromcomplete3Dshapesacquiredusingahigh-precisionscanningsystem:Interestforbodyweightestimation
初梦苑,刘刚,司永胜,等.基于三维重建的奶牛体重预估方法
CHUM,LIUG,SIY,etal.Predictingmethodofdairycowweightbasedonthree-dimensionalreconstruction
MEHMETK,BARDAKIOLUHE.Estimationofbodyweightandbodyconditionscoreindairycowsbydigitalimageanalysismethod
SUNX,HANG,YUM,etal.EstimationofgeneticparametersofbodysizeandbodyweightinHolsteincattleandtheircorrelationwithmilkproductionperformance
XIANGY,NAKAMURAS,TAMARIH,etal.3Dmodelgenerationofcattlebyshape-from-silhouettemethodforictagriculture
XAVIERC,LECOZLERY,DEPUILLEL,etal.Theuseof3-dimensionalimagingofHolsteincowstoestimatebodyweightandmonitorthecompositionofbodyweightchangethroughoutlactation
RUCHAYA,KOBERV,DOROFEEVK,etal.ComparativeanalysisofmachinelearningalgorithmsforpredictingliveweightofHerefordcows
杭孝.早期预警在奶牛疾病中的应用分析
季凤娟.奶牛常见疾病的防治措施
HORTETP,SEEGERSH.Lossinmilkyieldandrelatedcompositionchangesresultingfromclinicalmastitisindairycows
彭丹丹,陈健,赵越,等.脏污程度对奶牛乳区温度分布规律的研究
PENGD,CHENJ,ZHAOY,etal.Effectofsmudgydegreeontemperaturedistributionoftheuddersurfaceindairycow
VIGUIERC,ARORAS,GILMARTINN,etal.Mastitisdetection:Currenttrendsandfutureperspectives
MIRAQ,BANSALBK,GUPTADK.SubclinicalmastitisinmachinemilkeddairyfarmsinPunjab:Prevalence,distributionofbacteriaandcurrentantibiogram
SCHUKKENYH,WILSONDJ,WELCOMEF,etal.Monitoringudderhealthandmilkqualityusingsomaticcellcounts
MAUROZ,VERONICAR,FABIOL,etal.Firstevaluationofinfraredthermographyasatoolforthemonitoringofudderhealthstatusinfarmsofdairycows
蔡一欣,马丽,刘刚.奶牛隐性乳房炎便携式计算机视觉快速检测系统设计与试验
CAIY,MAL,LIUG.Designandexperimentofrapiddetectionsystemofcowsubclinicalmastitisbasedonportablecomputervisiontechnology
VOORTM,JENSEND,KAMPHUISC,etal.Invitedreview:Towardacommonlanguageindata-drivenmastitisdetectionresearch
LVENDAHLP,SRENSENLP.Frequentlyrecordedsensordatamaycorrectlyprovidehealthstatusofcowsifdataarehandledcarefullyanderrorsarefilteredaway
DENGZ,HOGEVEENH,LAMT,etal.Performanceofonlinesomaticcellcountestimationinautomaticmilkingsystems
JORGENSENCH,KRISTENSENAR,OSTERGAARDS,etal.Useofonlinemeasuresofl-lactatedehydrogenaseforclassificationofposttreatmentmammaryStaphylococcusaureusinfectionstatusindairycows
DALENG,RACHAHA,NORSTEBOH,etal.Thedetectionofintramammaryinfectionsusingonlinesomaticcellcounts
HOGEVEENH,KAMPHUISC,STEENEVELDW,etal.Sensorsandclinicalmastitis:Thequestfortheperfectalert
DOMINIAKKN,KRISTENSENAR.Prioritizingalarmsfromsensor-baseddetectionmodelsinlivestockproduction—Areviewonmodelperformanceandalarmreducingmethods
JONESBF,PLASSMANNP.Digitalinfraredthermalimagingofhumanskin
SATHIYABARATHIM,JEYAKUMARS,MANIMARANA,etal.InfraredthermographytomonitorbodyandudderskinsurfacetemperaturedifferencesinrelationtosubclinicalandclinicalmastitisconditioninKaranfries(bostaurus×bosindicus)crossbredcows
杨春合,顾宪红,曹正辉,等.奶牛左右乳区温度温差作为隐性乳房炎检测指标的可行性研究
YANGC,GUX,CAOZ,etal.Studyonpossibilityofleftandrightquarterskintemperaturedifferenceasadetectingindicatorforsubclinicalmastitisindairycows
郭艳娇,杨圣慧,迟宇,等.基于热红外图像的奶牛乳区温度分布与乳房炎识别方法
GUOY,YANGS,CHIY,etal.Recognizingmastitisusingtemperaturedistributionfromthermalinfraredimagesincowudderregions
王彦超,康熙,李孟飞,等.基于改进YOLOv3-tiny的奶牛乳房炎自动检测方法
WANGY,KANGX,LIM,etal.AutomaticdetectionmethodfordairycowmastitisbasedonimprovedYOLOv3-tiny
HOVINENM,SIIVONENJ,TAPONENS,etal.Detectionofclinicalmastitiswiththehelpofathermalcamera
WOLLOWSKIL,BERTULATS,KOSSATZA,etal.Shortcommunication:Diagnosisandclassificationofclinicalandsubclinicalmastitisutilizingadynamometerandahandheldinfraredthermometer
陈丽媛,洪小华,颜培实.我国南方冬季和夏季肉牛体感温度研究
CHENL,HONGX,YANP.EffectivetemperatureequationofcowsduringwinterandsummerseasonsinsouthernofChina
FRANZEU,GEIDELS,HEYDEU,etal.Investigationofinfraredthermographyforautomatichealthmonitoringindairycows
BERRYRJ,KENNEDYAD,SCOTTSL,etal.Dailyvariationintheuddersurfacetemperatureofdairycowsmeasuredbyinfraredthermography:Potentialformastitisdetection
POURSABERIA,BAHRC,PLUKA,etal.Real-timeautomaticlamenessdetectionbasedonbackpostureextractionindairycattle:Shapeanalysisofcowwithimageprocessingtechniques
TADICHN,FLORE,GREENL.Associationsbetweenhooflesionsandlocomotionscorein1098unsounddairycows
SJSTRMK,FALLN,BLANCO-PENEDOI,etal.LamenessprevalenceandriskfactorsinorganicdairyherdsinfourEuropeancountries
GRIMMK,HAIDNB,ERHARDM,etal.Newinsightsintotheassociationbetweenlameness,behavior,andperformanceinSimmentalcows
AFONSOJS,BRUCEM,KEATINGP,etal.ProfilingdetectionandclassificationoflamenessmethodsinBritishdairycattleresearch:Asystematicreviewandmeta-analysis
SPRECHERDJ,HOSTETLERDE,KANEENEJB.Alamenessscoringsystemthatusespostureandgaittopredictdairycattlereproductiveperformance
SCHLAGETER-TELLOA,BOKKERSEAM,KOERKAMPPWGG,etal.Relationbetweenobservedlocomotiontraitsandlocomotionscoreindairycows
ALSAAODM,FADULM,STEINERA.Automaticlamenessdetectionincattle
PIETTED,NORTONT,EXADAKTYLOSV,etal.Individualisedautomatedlamenessdetectionindairycowsandtheimpactofhistoricalwindowlengthonalgorithmperformance
宋怀波,姜波,吴倩,等.基于头颈部轮廓拟合直线斜率特征的奶牛跛行检测方法
SONGH,JIANGB,WUQ,etal.Detectionofdairycowlamenessbasedonfittinglineslopefeatureofheadandneckoutline
KANGX,ZHANGXD,LIUG.Accuratedetectionoflamenessindairycattlewithcomputervision:Anewandindividualizeddetectionstrategybasedontheanalysisofthesupportingphase
JIANGB,SONGH,WANGH,etal.Dairycowlamenessdetectionusingabackcurvaturefeature
KANGX,LIS,LIQ,etal.Dimension-reducedspatiotemporalnetworkforlamenessdetectionindairycows
VIAZZIS,BAHRC,HERTEMTVAN,etal.Comparisonofathree-dimensionalandtwo-dimensionalcamerasystemforautomatedmeasurementofbackpostureindairycows
JABBARKA,HANSENMF,SMITHML,etal.Earlyandnon-intrusivelamenessdetectionindairycowsusing3-dimensionalvideo
ALSAAODM,BüSCHERW.Detectionofhooflesionsusingdigitalinfraredthermographyindairycows
ALSAAODM,SYRINGC,DIETRICHJ,etal.Afieldtrialofinfraredthermographyasanon-invasivediagnostictoolforearlydetectionofdigitaldermatitisindairycows
康熙,李树东,张旭东,等.基于热红外视频的奶牛跛行运动特征提取与检测
KANGX,LIS,ZHANGX,etal.Featuresextractionanddetectionofcowlamenessmovementbasedonthermalinfraredvideos
FLOWERFC,SANDERSONDJ,WEARYDM.Hoofpathologiesinfluencekinematicmeasuresofdairycowgait
RUTTENCJ,VELTHUISAGJ,STEENEVELDW,etal.Invitedreview:Sensorstosupporthealthmanagementondairyfarms
JIANGB,WUQ,YINX,etal.FLYOLOv3deeplearningforkeypartsofdairycowbodydetection
OKURAF,IKUMAS,MAKIHARAY,etal.RGB-Dvideo-basedindividualidentificationofdairycowsusinggaitandtextureanalyses
BEGGSDS,JONGMANEC,HEMSWORTHPH,etal.LamecowsonAustraliandairyfarms:Acomparisonoffarmer-identifiedlamenessandformallamenessscoring,andthepositionoflamecowswithinthemilkingorder
VIAZZIS,BAHRC,SCHLAGETER-TELLOA,etal.Analysisofindividualclassificationoflamenessusingautomaticmeasurementofbackpostureindairycattle
WEIGELEHC,GYGAXL,STEINERA,etal.Moderatelamenessleadstomarkedbehavioralchangesindairycows
JONESBW.Behavioralgaitchangecharacterizationanddetectionusingprecisiondairymonitoringtechnologies
NUFFELAVAN,ZWERTVAEGHERI,WEYENBERGSVAN,etal.Lamenessde-tectionindairycows:Part2
BERCKMANSD,BAHRC,LEROYT,etal.Automaticdetectionoflamenessindairycattle—Analyzingimageparametersrelatedtolameness
SONGX,LEROYT,VRANKENE,etal.Automaticdetectionoflamenessindairycattle—Vision-basedtrackwayanalysisincow'slocomotion
PLUKA,BAHRC,POURSABERIA,etal.Automaticmeasurementoftouchandreleaseanglesofthefetlockjointforlamenessdetectionindairycattleusingvisiontechniques
POURSABERIA,BAHRC,PLUKA,etal.Onlinelamenessdetectionindairycattleusingbodymovementpattern(BMP)
ZHAOK,BEWLEYJM,HED,etal.Automaticlamenessdetectionindairycattlebasedonlegswinganalysiswithanimageprocessingtechnique
JIANGB,SONGH,HED.Lamenessdetectionofdairycowsbasedonadoublenormalbackgroundstatisticalmodel
康熙,张旭东,刘刚,等.基于机器视觉的跛行奶牛牛蹄定位方法
KANGX,ZHANGX,LIUG,etal.Hooflocationmethodoflamedairycowsbasedonmachinevision
SCHLAGETER-TELLOA,HERTEMTVAN,BOKKERSEAM,etal.Performanceofhumanobserversandanautomatic3-dimensionalcomputer-vision-basedlocomotionscoringmethodtodetectlamenessandhooflesionsindairycows
O'LEARYNW,BYRNEDT,O'CONNORAH,etal.Invitedreview:Cattlelamenessdetectionwithaccelerometers
张珂,吴志明,闫若潜,等.奶牛养殖场生物安全体系建设的现状、问题及对策
ZHANGK,WUM,YANR,etal.Currentsituation,problemsandcountermeasuresforconstructionofbiosecuritysystemindairyfarms
闫河,陈晓暾.我国电子信息制造业出厂价格指数特征探究
KANGX,ZHANGX,LIUG.Areview:Developmentofcomputervision-basedlamenessdetectionfordairycowsanddiscussionofthepracticalapplications
肩高、臀宽、体宽、坐骨...
肩高、臀宽、体宽、坐骨...基于点云采集设备的奶牛体尺指标测量22014...Researchesoncomputervision-basedmeasurementofbodysizeofdairycows
早期研究多利用传统的数字图像处理方法定位奶牛体尺测点，估测奶牛体尺参数［20］，但该方法易受环境等外界因素影响，导致奶牛体尺估测精度较低.随着人工智能技术的快速发展，深度学习技术应用于奶牛体尺测点定位研究，显著地提高了检测精度［26］.深度学习技术可以筛选质量较好的图像，用于定位体尺测点，在奶牛形态测量评估中具有较大的应用潜力［25］.基于二维图像的奶牛体尺测量研究处理算法较为简单，且计算速度快，基本满足奶牛体尺实时测量的要求.但对于奶牛胸围、腹围、管围等体尺围度的测量，则表现出一定的局限性....
早期研究多利用传统的数字图像处理方法定位奶牛体尺测点，估测奶牛体尺参数［20］，但该方法易受环境等外界因素影响，导致奶牛体尺估测精度较低.随着人工智能技术的快速发展，深度学习技术应用于奶牛体尺测点定位研究，显著地提高了检测精度［26］.深度学习技术可以筛选质量较好的图像，用于定位体尺测点，在奶牛形态测量评估中具有较大的应用潜力［25］.基于二维图像的奶牛体尺测量研究处理算法较为简单，且计算速度快，基本满足奶牛体尺实时测量的要求.但对于奶牛胸围、腹围、管围等体尺围度的测量，则表现出一定的局限性....Researchonbodymeasurementandweightestimationofcowsbasedon3Dreconstruction12020...为0.75
早期研究多利用传统的数字图像处理方法定位奶牛体尺测点，估测奶牛体尺参数［20］，但该方法易受环境等外界因素影响，导致奶牛体尺估测精度较低.随着人工智能技术的快速发展，深度学习技术应用于奶牛体尺测点定位研究，显著地提高了检测精度［26］.深度学习技术可以筛选质量较好的图像，用于定位体尺测点，在奶牛形态测量评估中具有较大的应用潜力［25］.基于二维图像的奶牛体尺测量研究处理算法较为简单，且计算速度快，基本满足奶牛体尺实时测量的要求.但对于奶牛胸围、腹围、管围等体尺围度的测量，则表现出一定的局限性....ApplicationofdepthsensortoestimatebodymassandmorphometricassessmentinNelloreheifers22021...为0.75
平均相对误差为3.67%，标准差...
Fotric-625c红外...
灵敏度为100.00%，特异性为75.00%，准确率...
注：①k最邻近分类算法（K-NearestNeighbor，KNN）；②双向长短词记忆模型（Bi-directionalLongShort-TermMemory，BiLSTM）；...
注：①k最邻近分类算法（K-NearestNeighbor，KNN）；②双向长短词记忆模型（Bi-directionalLongShort-TermMemory，BiLSTM）；...Featuresextractionanddetectionofcowlamenessmovementbasedonthermalinfraredvideos