Real-time automatic lameness detection based on back posture extraction in dairy cattle: Shape analysis of cow with image processing techniques

In this paper results on utilizing image analysis techniques towards early lameness detection in dairy cattle are presented. Data from two different dairy farms in Belgium were gathered. Preprocessing on raw data is required because of non-predictable behaviours of cows such as stopping for a while in front of the camera or non-uniform walking behaviour during experiments. Prelocalization of cow in each frame has been done based on two steps separation: (1) A coarse estimation of moving objects was obtained through background subtraction, (2) second statistical analysis of intensities in gray-scale image along with binarization was utilized to detect moving object in video. A common problem in on-farm collected videos is the similarity of the background and the cow's body colour since the use of classic algorithms for segmentation purposes does not work. Here a hierarchy background/foreground exaggeration is proposed to segment the cow in each frame and track it in video. The combination of logarithm and exponential, background subtraction as well as statistical filtering are used to find the accurate shape of the cow. Furthermore, the back posture of each cow during standing and walking was extracted automatically. It was done by detecting the arc of back posture and fitting a circle through selected points on the spine line. The average inverse radius of four frames displaying the hind hoofs in contact with the ground (two frames for each hoof in a row) was assigned to the cow. Based on this curvature value, a score representing the status of lameness in the individual cow was given automatically. Experimental results from two different databases show promising results in automatic lameness detection based on back posture information.     

新疆地区荷斯坦牛长寿性与产奶量的遗传力估计及相关性分析

【目的】产奶性状和长寿性状均是奶牛育种中的重要性状。近年来奶牛选育中追求高产,在产奶量不断提高的同时,长寿性却呈下降趋势。随着平衡育种理念的发展,各奶业发达国家陆续将长寿性状纳入其奶牛综合选择指数。为此,探究产奶量和奶牛长寿性间的关系,为新疆地区制定中国荷斯坦奶牛综合选择指数,实现平衡育种和奶业高质量健康发展提供借鉴。【方法】收集新疆地区3个规模化牛场1997—2020年间的各项生产记录,计算在群天数和生产寿命,共计获得7 206条奶牛长寿性记录和15 218条头胎305 d产奶量记录。收集个体三代系谱信息共计18 183条,其中包括903头公牛以及20 883头母牛。首先利用SAS 9.2软件的GLM过程分析场、出生年份、出生季节和初产月龄对新疆地区中国荷斯坦牛在群天数、生产寿命和头胎305 d产奶量的影响,计算出在群天数、生产寿命和头胎305 d产奶量的最小二乘均值。再利用SPSS 19.0 软件计算新疆地区中国荷斯坦牛在群天数、生产寿命和头胎305 d产奶量Pearson相关系数。其次利用DMU软件中AI-REML结合EM算法并配合多性状动物模型,估计新疆地区荷斯坦奶牛在群天数、生产寿命和头胎305 d产奶量的遗传方差、表型方差、协方差,计算在群天数、生产寿命和头胎305 d产奶量之间的遗传相关。最后利用动物模型 BLUP 法对新疆地区中国荷斯坦牛在群天数、生产寿命和头胎305 d产奶量进行育种值的估计,并绘制遗传趋势图。【结果】对新疆地区中国荷斯坦牛在群天数、生产寿命和头胎305 d产奶量进行基本统计分析结果显示：新疆地区中国荷斯坦奶牛平均在群天数1 754.7 d,平均生产寿命937.33 d,头胎305 d产奶量平均为9 362.94 kg。GLM程序分析显示不同场、出生年份、出生季节以及初产月龄效应对在群天数、生产寿命以及头胎305 d产奶量均有极显著影响（P<0.01）。新疆地区中国荷斯坦牛在群天数、生产寿命和头胎305 d产奶量的遗传力为0.11（0.03）、0.11（0.03）和0.33（0.03）。在群天数和生产寿命之间遗传和表型相关高,相关系数分别为0.99和0.98;305 d产奶量与在群天数和生产寿命表型之间呈正相关,相关系数分别为0.079和0.077,而遗传相关则呈负相关,相关系数分别为-0.18和-0.20。对新疆地区中国荷斯坦在群天数、生产寿命和头胎305 d产奶量遗传趋势图绘制可知,在群天数和生产寿命,均随着参测牛头数增加育种值变化较大,整体呈下降趋势;荷斯坦牛的头胎305 d产奶量同样随着参测牛头数的增加,育种值变化较大,但是总体呈现上升趋势。【结论】因此在对产奶量进行选育提高同时,应当兼顾长寿性,避免随着产奶量的提高,导致奶牛长寿性的下降。     

奶牛蹄病与矿物元素相关性研究

[目的]探究奶牛体内矿物元素含量与蹄病的关系。[方法]从某大型集约化奶牛场随机抽取8头患蹄病的奶牛和8头健康奶牛,采用原子吸收分光光度法检测患蹄病奶牛与健康奶牛血清中一些矿物元素(K、Fe、Cu、Mg、Mn、Na、Zn、Ca)的含量,分析了矿物元素的代谢与奶牛蹄病之间的关系。[结果]与对照组相比,试验组奶牛血清中Zn和Mn含量差异极显著(P0.01),试验组Ca和Mg含量差异显著(P0.05),试验组K、Fe、Cu和Na含量差异不显著(P0.05)。[结论]患蹄病奶牛内矿物元素代谢异常,特别是与骨骼生长发育有关的元素代谢紊乱,容易导致蹄病的发生。     

Automatic blood sampling in dairy cows

Loose housing systems for dairy cows are becoming increasingly common, raising new questions in the area of animal health and welfare. Some of these questions can be addressed by studying the variation in blood parameters, such as glucocortistioids. However, the traditional manual blood sampling procedure can in itself affect the responses of the animal. To address this issue we have developed a device for automated collection of multiple blood samples. The device is placed on the back of the animal and allows the animal to be kept in all types of environments, either alone or in social groups. The animal can move freely: no restraints and no handling of the animal are necessary during blood sampling.Three experiments were conducted to study the performance of the system and how the blood sampling procedure affected the cortisol responses in dairy cows. In the first experiment the accuracy of sample size and timing were investigated.In the second experiment, automatic samples were collected from six cows kept in tie-stalls, with samples taken at 3-min intervals for 38 min. At 12, 24, 36 and 38 min, a manual sample was taken by vein puncture. None of the cows showed a consistently increased cortisol response to automatic sampling, while two cows showed increased cortisol concentration after vein puncture.In the third experiment automatic blood sampling was conducted with 12 cows kept in a loose housing system. In the morning, over a period of 1.5 h, 14 blood samples were taken at varying time intervals from each cow. In the afternoon, over a period of 2.5 h, a further 14 samples were taken. Cortisol concentrations in the blood samples suggest that the cows were not affected by the blood sampling procedure.On some occasions during these experiments, it was not possible to get a sample. For the manual vein puncture 3 out of 24 samples (12.5%) were not successfully taken within a time limit of 2 min. During the automatic blood sampling process, the catheter occasionally became twisted and blocked by the movement of the cow's head. This occurred in 8 out of 84 samples (9.5%). In four of the manual samples, considerable hemolysis had occurred during the sampling procedure. No hemolysis was observed in the automatic blood samples.Automated blood sampling offers the opportunity to collect a series of samples without disturbing the animal. Data suggest that automatic blood sampling is a superior alternative to manual blood sampling, even in a tie barn.     

不同养殖模式及存栏规模牛场奶牛生产性能分析

对山东省86家奶牛群体改良(DHI)参测牛场的生产性能数据进行了综合统计,采用最小二乘分析法探究了养殖模式(规模化牧场和小区)及饲养规模对牛只生产性能是否存在影响,同时比较了近年来不同规模奶牛场生产性能变化情况。结果表明,山东省DHI参测奶牛场以规模化牧场为主,牧场的存栏规模以300～600头最多,其次为300头以下、600～1000头、1000头以上。不同养殖模式和养殖规模牛场之间在各项生产性能指标上存在显著差异(P<0.05):牧场养殖模式下牛群平均胎次、生奶产量和质量水平显著高于小区模式;不同养殖规模牛场生产性能比较显示,日产奶量、校正奶量和305天奶量均以300头以下牛场为最高(P<0.05),乳脂率以1000头以上牛场为最高(P<0.05),乳蛋白率以600～1000头存栏规模的牛场最高(P<0.05)。从近7年校正奶量变化情况看,存栏600～1000头的奶牛场校正奶量上升趋势最为明显。     

Ca、P营养失衡对乳牛蹄病的影响

为了探讨钙(Ca)、磷(p) 营养失衡对乳牛蹄病的影响, 我们对太原A牧场(蹄病多发场) 与太原B牧场(蹄病少发场) 的健康牛与蹄病牛的血清、日粮及牧场的土壤、水源中Ca、p含量进行测定分析。结果表明,Ca、p比例失调, 环境土壤中钙含量偏高, 饮用水水质过硬是乳牛蹄病的诱因。因此合理搭配日粮, 调整日粮中Ca、P比, 改善环境土壤及水源状况对本病的防治有重要意义。     

Study of Digital Management System of Milking Process on Large-Sized Dairy Farm

This study was to supply the systemic and full milking process data to support the implementation of both dairy herd improvement （DHI） and digital feeding of dairy cattle. This study designed the relational structured database and developed a set of digital management information system on milking process of intensive dairy farm using Visual Basic 6.0, Access databases, and Crystal report combining the milking characteristics of a grown cow, such as quality and sanitation testing indexes of raw milk. The system supplies a series of convenient, intelligent input interfaces of crude datum, and can count, analyze, and graphically show milking datum based on different types and different parities of cows or herds in a specific duration, and can dynamically produce some important derived data, such as days of grown cow, daily average of milk production of grown cow, days of cow milk production, and daily average of milking cow production; and can carry out all-pervasive data mining. With the help of system analysis and software design techniques, it is possible to realize precision farming for a dairy cattle herd based on whole digital management of milking process and realtime prediction on nutrient requirements and ration of dairy cattle, as well as dairy herd improvement.     

中草药添加剂对不同泌乳阶段奶牛生产性能的影响

本文主要探讨了中草药添加剂对不同泌乳阶段奶牛生产性能的影响。选择胎次相同、产奶量相近的健康荷斯坦牛24头,根据泌乳期的天数分为4个区组,每组以试验前后自身作为对照;将中草药添加剂按1%的比例添加到基础日粮中,饲喂28d。结果表明,除泌乳131～158d组外,其他各试验区组奶牛产奶量都有不同程度的提高,特别对泌乳237～277d的奶牛,其增奶效果最优,产奶量与试验前相比提高了4.3%（P〈0.05）,且乳房炎和蹄病的发病率为0。这表明中草药添加剂可提高泌乳奶牛产奶量,延缓中后期奶牛泌乳的下降趋势,并能对抗热应激,减少乳房炎的发病率,提高经济效益。     

木薯渣对奶牛产奶量和奶品质的影响

试验选用30头体重约400kg、健康无病、胎次相同,且处于泌乳中期,产奶量相近（P〉0.05）的荷斯坦奶牛,随机分为2组：对照组10头,用产奶期正常饲料饲喂;试验组20头,每头牛在对照组日粮基础上投5kg木薯渣替代5kg苜蓿干草,研究木薯渣对奶牛产奶量及奶品质的影响。结果表明：在产奶期日粮中添加木薯渣对奶牛的产奶量没有显著影响（P〉0.05）;从乳成分分析统计数据看,其对奶牛的乳成分影响不明显,乳脂、乳蛋白、乳糖、乳固形物和非脂固形物均无显著差异（P〉0.05）。     

Detecting cow's lameness using force sensors

Our aim is to automatically detect cow's leg problems. The first system is a four-balance system, where each of the legs is weighed when a cow is in a milking robot. Its functionality is limited to milking robot and complications in interpreting results occur, since often not all of the legs are properly in each balance or the cow is leaning the robot. Therefore, we introduce a new system. It is a mat made of electromechanical film, Emfit, which can detect only dynamic forces. Its benefit is that its use is not limited to milking robot, but it can be set up in any corridor along which the cows walk. Preliminary tests with walking cows indicate that it has potentiality to separate lame cows from healthy cows by different force–time behaviour.     

Automatic real-time monitoring of locomotion and posture behaviour of pregnant cows prior to calving using online image analysis

Monitoring the locomotion and posture behaviour of pregnant cows close to calving is essential in determining if there is a need for human intervention to assist parturition. In this study an automatic real-time monitoring technique is described in detail which allows identifying the locomotion and posturing behaviour of pregnant cows prior to calving. For this purpose video surveillance images of eight cows for the last 24 h prior to their calving were analysed. Data on seven different variables with time were obtained for each cow using an automatic real-time monitor. These were namely: x–y coordinates of the geometrical top view centre point of the cow; walking trajectory; distance walked; orientation of the main axis; body width/length ratio; hip length and back area. These variables were then used to classify specific behaviours such as standing or lying (including incidences of motion during lying), and eating or drinking. On average 85% of the standing and lying and 87% of the eating or drinking behaviour of the eight cows during the last 24 h before calving could be correctly classified. However, the developed technique needs to be further validated with additional tests in the field.     

长沙地区奶牛养殖现状调查

为探明湖南奶牛养殖发展缓慢的原因，以促进湖南奶业的发展，对长沙地区主要奶牛场奶牛的产奶性能和饲料使用情况进行了调查．结果表明：1）季节对奶牛产奶量和乳成分有极显著影响，8月奶牛的产奶量降至最低；2）不同牛场间4～9月乳成分存在极显著差异，且测定月份和牛场对乳蛋白互作影响极显著；3）长沙奶牛粗饲料明显缺乏．据此，笔者提出了发展长沙（湖南）奶业的主要措施：1）立足本地资源，开发具有长沙（湖南）特色的奶牛饲料，如全株青贮饲料稻和整粒棉籽；2）采取综合措施缓解奶牛热应激；3）提高奶牛饲养管理水平．     

北方寒区奶牛高产高效益配套技术的研究

为了研究北方;寒区奶牛高产高效益配套技术;我们受农业部和农场总局委托,“八五”期间在８５１１农场,香坊农场,８５２农场,８５０和克山农场奶牛群,根据北方寒区特点,在奶牛选种,后备牛的培育,饲养管理,饲料选择和调制,配种和疾病防治等方面采取了全方位措施。     

北京荷斯坦牛体型性状相关性分析

本研究旨在评估北京市荷斯坦牛体型性状之间的相关性,为选种选配提供依据。试验对来自北京地区牛场的1 391头荷斯坦母牛的各部分体型性状进行相关性分析。结果显示乳房、乳用特征、肢蹄分别与体型总分之间的正相关性较强（P<0.01）。同时肢蹄、乳房分别与乳用特征之间呈较强的正相关性（P<0.01）,可重点改良这三个特征性状以提高体型总分。此外,体躯大小、胸宽、体深分别与结构容量之间呈极强正相关（P<0.05）,且体躯大小、胸宽分别与体深之间呈极强正相关（P<0.05）;蹄角度与骨质地之间、蹄踵深度与后肢侧视之间正相关性极强（P<0.01）;前乳房附着与后附着高度之间、前乳头位置与后附着高度之间、前乳头位置与后乳房宽度之间均呈极强负相关（P<0.05）,而乳房深度、悬韧带分别与后乳头位置呈极强正相关（P<0.05）。相关性强的性状中针对性选择误差小、易测量的性状,重视体型性状和生产性状的同步优化,有助于北京地区荷斯坦牛的品种改良。     

秋季奶牛腐蹄病高发的原因及综合防治

奶牛腐蹄病是目前危害农村散养户奶牛生产、导致奶牛生产性能下降的重要疾病之一。该病发生后,蹄的真皮和角质层组织发生化脓性病理变化,其特征是真皮坏死与化脓,角质溶解,病牛疼痛,跛行,严重影响奶牛生产性能。特别在秋季,更要加强对该病的防治。     

宁夏中地生态牧场奶牛淘汰原因与胎次和月份的关系研究

对宁夏中地生态牧场奶牛淘汰原因与胎次和月份关系进行了研究,结果发现在2015年和2016年中,乳腺疾病、肢蹄病和消化系统疾病均是引起贺兰中地生态牧场奶牛淘汰的主要因素;淘汰率最高的月份均为5月份,2015年第2胎的淘汰率最高,2016年第1胎的淘汰率最高.奶牛的淘汰原因与胎次和月份均差异不具有统计学意义,但淘汰原因与月份有一定的相关性.     

集约化畜牧业生产与奶牛福利

讨论了集约化畜牧业生产与奶牛福利的关系,提出由集约化畜牧业生产对奶牛福利的影响,并从建立奶牛安全生产环境、改进奶牛生产工艺、奶牛福利法规建设等方面加强奶牛福利建设进行了设想。     

不同牛场奶牛泌乳曲线拟合及参数分析

以3个牛场2009～2010年参与DHI测定的成年母牛的泌乳期数据为资料,用Wood模型通过SPSS的非线性参数和决定系数来拟合泌乳曲线并分析泌乳曲线参数在不同牛场和胎次的差异.结果表明,不同牛场估测的泌乳曲线模型参数存在显著的差异(P＜0.05),导致部分泌乳特性参数也有一定的差异.这些差异可能是牛群选育或饲养管理或二者共同的结果.因此,分析泌乳曲线应与牛场自身的饲养管理和选育程度相结合.     

基于高通量测序技术分析奶牛乳房炎关联微生物群落结构及多样性

[目的]研究奶牛乳房炎主要病原微生物及乳汁中微生物群系分布.[方法]对CJ和HZS 2个牛场的奶牛进行乳房炎检查,视觉和触觉等临床检查奶牛乳房及乳汁后,将2个场中确诊为临床乳房炎的奶牛随机采集乳汁样本各3份,并将非临床型奶牛的乳汁样本随机采集各1份,通过对乳房炎奶牛和非临床型奶牛乳样中细菌的16S rRNA基因V3-V...     

规模化奶牛场数字化网络管理平台开发与应用

【目的】利用网络数据库技术,建立规模化奶牛场生产过程数据网络整合与智能分析共享平台,以满足规模化奶牛场数字化管理水平,提高繁育效率及遗传进展。【方法】基于奶牛从发情、配种、孕检、妊娠、产犊、泌乳、干奶到下一个繁殖周期的生产管理流程,将奶牛划分为犊牛、青年牛、育肥牛、泌乳牛、干奶牛5种类型,在设计采集生产母牛及公牛基本信息及数据标准规范后,采用Microsoft.Net框架、SQL Server 2008网络数据库及FusionCharts网络绘图等技术,研究构建规模化奶牛场繁殖、泌乳及健康等数据管理及智能分析平台。【结果】平台一级子系统包括系统维护、牛只管理、繁殖管理、产乳管理、饲养管理、健康管理、统计分析及场内管理。在8大子系统上总共实现的功能模块多达96项。其中,系统维护6项,牛只管理10项,繁殖管理13项,含DHI数据的产乳管理14项,饲料管理4项,健康管理10项,统计分析18项以及场内管理21项。平台主要实现了对奶牛繁殖与泌乳生产核心数据的网络远程数字化管理,主要包括公、母牛个体繁殖状态的全程记录、不同胎次的泌乳数据的记录、牛只资料卡等在线动态分析、各类基础数据的排序与输出等。在数据挖掘分析上,可在线计算平均胎间距、泌乳牛胎次结构、高低产母牛数量、系谱跟踪及近交系数等。在生产提示功能设计上,可处理的项目包括发情配种、初检复检、产犊泌乳、干奶、催乳、淘汰及犊牛断奶等各类将发生的生产事件,而且可对大部分统计分析数据进行可视化图形渲染处理,如胎次产量分布图、年度产奶比较图、二叉树谱系图及泌乳曲线图等。【结论】对生产过程基本数据的分析与数据挖掘,提升了基础数据的利用价值,更有利于奶牛场管理者的科学决策。