Behavior Of Cows Research Articles

Interpolation Methods to Improve Data Quality of Indoor Positioning Data for Dairy Cattle

Position data from real-time indoor positioning systems are increasingly used for studying individual cow behavior and social behavior in dairy herds. However, missing data challenges achieving reliable continuous activity monitoring and behavior studies. This study investigates the pattern of missing data and alternative interpolation methods in ultra-wideband based real-time indoor positioning systems in a free-stall barn. We collected 3 months of position data from a Swedish farm with around 200 cows. Data sampled for 6 days from 69 cows were used in subsequent analyzes to determine the location and duration of missing data. Data from 20 cows with the most reliable tags were selected to compare the effects of four different interpolation methods (previous, linear interpolation, cubic spline data interpolation and modified Akima interpolation). By comparing the observed data with the interpolations of the simulated missing data, the mean error distance varied from around 55 cm, using the previously last observed position, to around 17 cm for modified Akima. Modified Akima interpolation has the lowest error distance for all investigated activities (rest, walking, standing, feeding). Larger error distances were found in areas where the cows walk and turn, such as the corner between feeding and cubicles. Modified Akima interpolation is expected to be useful in the subsequent analyses of data gathered using real-time indoor positioning systems.

Acoustic sensors for detecting cow behaviour

Acoustic technologies provide a non-invasive method to generate information about the health, welfare, and environmental impact of livestock. This study demonstrated that rear leg attached acoustic sensors can be used to differentiate between seven different acoustic classes, with six based on cow behaviours (Grazing, Breathing, Walking, Lying Down, Dung, Vocalization, Other) that were obtained from more than 150 cows under grazing conditions. The overall accuracy of the ensemble classification model was 96.2% based on a total of 700 acoustic recordings. The performance of the models for the duration, frequency (number of events per 10 seconds), or period (average time between events) of the six animal behaviours had an average coefficient of determination of R2 = 0.93. The model for respiration rate (Breathing class) while sleeping (R2 = 0.99) provides an alternative to more invasive differential pressure and thermistor-based methods. The model performance for bite rate (R2 = 0.91) was consistent with results obtained previously with collar and forehead attached microphones. There was ten-fold variation in the duration of dung events and the model for the duration of dung events (R2 = 0.96) allows for estimation of the total amount of dung deposited per day. The acoustic technology also provides (R2 = 0.91) an alternative to accelerometer-based methods for stepping frequency. Lying down events were characterised by scratching sounds generated by the microphone rubbing against the pasture that provided good prediction of duration of the time to lie down (R2 = 0.86). Models for vocalization duration (R2 = 0.92) and classification of the vocalization class (sensitivity 0.99; precision 0.95) demonstrate the feasibility of acoustic-based determination of vocalization traits, which provide information on the welfare and state of the animal.

Effects of concentrate allowance and individual dairy cow personality traits on behavior and production of dairy cows milked in a free-traffic automated milking system

The primary objective of this study was to determine whether the level of concentrate allowance in an automated milking system (AMS) affects the feed intake, eating behavior, milking activity, and performance of lactating dairy cows. The secondary objective of this study was to describe how the response to concentrate allocation, specifically in feeding and milking behavior, varies with cow personality traits. Fifteen Holstein cows were used in a crossover design with two 28-d periods, each including 14 d of adaptation and 14 d of data collection. The cows were housed in a freestall pen with free-traffic access to the AMS. Treatments consisted of a basal partial mixed ration (PMR) common to both treatment groups, with a concentrate allowance (on dry matter basis) of (1) 3.0 kg/d in the AMS (L-AMS) or (2) 6.0 kg/d in the AMS (H-AMS). Between the 2 treatment periods, each cow was assessed for personality traits using a combined arena test consisting of exposure to a novel environment, novel object, and novel human. Principal component analysis of behaviors observed during the novel environment and object tests revealed 3 factors (interpreted as active, social, and alert-curious) that together explained 76% of the variance, whereas principal component analysis of the novel human test revealed 2 factors (interpreted as active-vocal and fearful of novel humans) that together explained 77% of the variance. When on the H-AMS treatment, PMR dry matter intake (DMI) was less (24.5 vs. 26.0 kg/d) and AMS concentrate delivery was greater (5.9 vs. 3.1 kg/d), as per design. Consequently, total DMI was greater on the H-AMS treatment (30.4 vs. 29.1 kg/d). When on the H-AMS treatment, cows who were more alert-curious consumed more PMR, whereas cows who were more fearful of the novel human were less likely to receive the maximum amount of AMS concentrate available, limiting their total DMI and increasing the day-to-day variability of that intake. Although this was a preliminary study, these data suggest an association between dairy cow personality traits and how cows respond to increased AMS concentrate allowance.

The effect of hides and parity on behavior of periparturient dairy cows at pasture

Under natural conditions, most parturient cows seek visual cover or seclude themselves from the herd when calving becomes imminent. Studies on calving site selection of dairy cows housed indoors show that predominantly older and dominant cows calve in secluded areas. This study aimed to investigate whether cows distance themselves further from herd members under spacious outdoor conditions and whether artificial hides on a pasture motivate cows to seek visual isolation from the herd. One hundred eighty-two Danish Holstein cows were allocated to 1 of 13 groups of 14 cows according to expected calving date. In 2 paddocks (each 75 × 150 m), 5 zones (75 m × 30 m) were defined; zone 0 contained the feeding and shade area, and zone 4 bordered an area with trees. Weekly and according to calving date, a group of 14 cows was moved either to a paddock that had 12 hides distributed evenly throughout zones 1 to 4 (7 groups) or to a paddock without hides (6 groups). The hides were wooden structures (3.3 m long × 1.0 m wide × 1.2 m high) that the cows could hide behind. One hundred seven cows calved within 14 d of having been moved to the paddock, and 82 of these (38 primiparous and 44 multiparous), and their calves, were included in the study. From 3 h before until 3 h after calving, the location, posture, and behavior of cows were recorded continuously from video recordings. In addition, location, posture, and behavior of the calves were recorded for the first 3 h after birth. Continuous variables were analyzed by linear mixed effects models, whereas binary data were analyzed using mixed effects logistic regression. The presence of hides did not affect selection of calving location in the paddock, but more primiparous than multiparous cows calved in zone 4; that is, furthest away from the feeding area. Before calving, primiparous cows were more likely to be positioned in zone 4 and less likely to approach and contact herd members. After calving, primiparous cows and their calves tended to be more likely to be positioned in zone 4. The more cows present in the paddock, the less likely the cows were to isolate (i.e., be further than 30 m away from any other cow) at the time of calving. In addition, the more cows in the paddock when a calf was born, the less time these calves spent in a lying posture. The results show that primiparous cows move further away from herd members at calving than multiparous cows and suggest that primiparous cows prefer to isolate themselves through distance rather than seeking artificial cover. Contrary to expectation, the hides did not make more cows select a calving site away from the feeding area.

Basic motion behaviour recognition of dairy cows based on skeleton and hybrid convolution algorithms

Accurate and rapid recognition of the basic motion behaviours of dairy cows is the key to intelligent perception of their health status. As spatiotemporal data with a long time range, a 3D convolution kernel is more suitable for feature extraction of dairy cows’ basic motion behaviours. Based on image features, the traditional 3D convolutional neural network (CNN) requires many parameters and insufficient depth, and the robustness is always poor. To accurately recognize basic motion behaviours (walking, standing, and lying) of cows, in this research, basic motion behaviours based on cow skeletons and a mixed convolution algorithm were proposed. The depth of the 3D CNN was increased by connecting a deep 2D convolution in series connection after each 3D convolution, and a parallel 2D convolution was added on the basis of the series. Then, the 3D and 2D feature maps were correlated to share spatial information. Simultaneously, the key point information of the cow’s skeleton corresponding to the frame was added in the form of a heat map in the parallel 2D convolution feature. While increasing the depth of the 3D convolutional network, it effectively controlled the number of model parameters and robustness. Three hundred cow videos containing the three specific motion behaviours were selected for testing. The results showed that after 5-fold cross validation, the final classification ACC of this method was 91.80%, which was 3.40% higher than that of a mixed 3D/2D convolutional tube (MiCT). To verify the robustness of the method, a gamma transform was applied to adjust the image brightness to simulate real brightness changes. Under different brightness, the accuracy of this method had a maximum offset of 6.40%, which was significantly lower than that of temporal segment networks (TSNs) and MiCT. Furthermore, the classification ACC fluctuated slightly even when adding different degrees of random noise to the cow skeleton. All of the results showed that the proposed method was effective for the classification of walking, standing, and lying behaviours of cows and can be used to identify the basic motion behaviours of cows.

Effect of dietary protein level and corn processing on behavior activity of high producing dairy cows

The objective of this experiment was to evaluate the effects of corn processing and protein level on the feeding, lying, and post milking standing (PMS) behavior in high producing cows. Eight Holstein cows were randomly assigned to diets containing either finely ground (FGC) or steam flaked (SFC) corn based on either low (LP) or high (HP) protein content. Cows receiving LP had lower milk yield than cows receiving HP with similar DMI. Moreover, FGC-fed cows had higher DMI than SFC-fed cows with similar milk yields. Eating and rumination time tended to be lower and chewing time was lower in HP-fed cows than LP-fed cows. Cows fed SFC tended to have higher laying rumination interval and lower lying rumination bouts than cows fed FGC. Total and average PMS were lower in cows fed HP than LP. Cows fed LP had higher chewing activity in the daytime than cows fed HP. Our results suggested that the protein level and corn processing affect the standing and lying behavior of high producing dairy cows, although, this effect is marginal. Results also indicated that probably any change in the diet that increases the rumination and eating times could also improve the PMSThe objective of this experiment was to evaluate the effects of corn processing and protein level on the feeding, lying, and post milking standing (PMS) behavior in high producing cows. Eight Holstein cows were randomly assigned to diets containing either finely ground (FGC) or steam flaked (SFC) corn based on either low (LP) or high (HP) protein content. Cows receiving LP had lower milk yield than cows receiving HP with similar DMI. Moreover, FGC-fed cows had higher DMI than SFC-fed cows with similar milk yields. Eating and rumination time tended to be lower and chewing time was lower in HP-fed cows than LP-fed cows. Cows fed SFC tended to have higher laying rumination interval and lower lying rumination bouts than cows fed FGC. Total and average PMS were lower in cows fed HP than LP. Cows fed LP had higher chewing activity in the daytime than cows fed HP. Our results suggested that the protein level and corn processing affect the standing and lying behavior of high producing dairy cows, although, this effect is marginal. Results also indicated that probably any change in the diet that increases the rumination and eating times could also improve the PMS

A deep learning-based cow behavior recognition scheme for improving cattle behavior modeling in smart farming

Farming and animal husbandry applications are improvised with the implication of machine learning and artificial intelligence in recent years. The precise estimation, recommendations, and performances are the prime reason for the technology implication. Owing to the modern agricultural and animal cultures, this article introduces an innovative Behavior Recognition and Computation Scheme (BRCS) for predicting cow behaviors. The information from the swallowed microchip is processed based on the observed animal action that is used for the forecast. Considering the information to be rectilinear, the distractions and distribution patterns (data) are augmented in identifying and forecasting its behavior. The proposed scheme identifies the patterns using a deep recurrent learning paradigm recurrently. This pattern is distinguished for idle and non-idle observations for improving the prediction accuracy. Distinguished data patterns are mapped for the consecutive time and observation data in classifying abnormalities. The proposed scheme's performance is validated using the metrics accuracy, precision, computing time, and mean error.

Waiting time in the premilking holding pen and subsequent lying and walking behaviors of Holstein cows

The objective of this study was to assess the effect of waiting time (WT) in the premilking holding pen on the subsequent lying and walking behaviors of Holstein cows. A total of 108 cows (multiparous n = 95; primiparous n = 13), milked 3 times/d [AM (0700 h), PM (1500 h), and nighttime (2300 h)] were monitored for individual WT, which was defined as the time elapsed between the entrance of the first cow to the milking stall in a rotary milking system and the entrance of each subsequent cow housed in the same milking pen. Data for lying time (min), standing time (min), lying bouts (no.), and walking behavior (no. of steps) in 15-min intervals were collected using a commercial 3-dimensional accelerometer affixed on the hind left leg of each cow from early to mid lactation. Mean values (min/h, no./h) for each behavior calculated between consecutive milkings were compared among 3 WT categories: cows with WT ≤30 min (WT1), WT 30 to 60 min (WT2), and WT >60 min (WT3). The overall mean (SD) of WT during the study period was 31.2 (23.4) min, and the distribution of WT1, WT2, and WT3 in all cows through the monitoring period was 50.9, 43.3, and 5.7%, respectively. No significant effects were established for overall WT on any of the behaviors under analysis. However, subsequent analysis by milking shift indicated that, during the period following the night milking, cows in WT1 exhibited the shortest lying time, whereas cows in WT3 had the greatest number of lying bouts. No effect of WT category was observed on the number of steps after any of the milking shifts. Results of this study indicate a moderate effect of WT on lying behavior after the nighttime milking. Variables such as temperature-humidity index, parity, and time of day affected cow behavior and should be considered when evaluating the impact of routine management practices, such as milking, on commercial farms.

Automatic Detection Method of Dairy Cow Feeding Behaviour Based on YOLO Improved Model and Edge Computing.

The feeding behaviour of cows is an essential sign of their health in dairy farming. For the impression of cow health status, precise and quick assessment of cow feeding behaviour is critical. This research presents a method for monitoring dairy cow feeding behaviour utilizing edge computing and deep learning algorithms based on the characteristics of dairy cow feeding behaviour. Images of cow feeding behaviour were captured and processed in real time using an edge computing device. A DenseResNet-You Only Look Once (DRN-YOLO) deep learning method was presented to address the difficulties of existing cow feeding behaviour detection algorithms’ low accuracy and sensitivity to the open farm environment. The deep learning and feature extraction enhancement of the model was improved by replacing the CSPDarknet backbone network with the self-designed DRNet backbone network based on the YOLOv4 algorithm using multiple feature scales and the Spatial Pyramid Pooling (SPP) structure to enrich the scale semantic feature interactions, finally achieving the recognition of cow feeding behaviour in the farm feeding environment. The experimental results showed that DRN-YOLO improved the accuracy, recall, and mAP by 1.70%, 1.82%, and 0.97%, respectively, compared to YOLOv4. The research results can effectively solve the problems of low recognition accuracy and insufficient feature extraction in the analysis of dairy cow feeding behaviour by traditional methods in complex breeding environments, and at the same time provide an important reference for the realization of intelligent animal husbandry and precision breeding.

Supplementation of a dried, fungal fermentation product with fibrolytic enzymatic activity in the diet of dairy cows on feeding behavior, metabolic profile, milk yield, and milk composition

• Increased milk production from high-performance cows and decreased somatic cell count Supplementation with exogenous fibrolytic enzymes in the diet of ruminants can improve the use of nutrients by animals and, consequently, improve their productive performance. The aim of this study was to evaluate the effect of supplementing a dried fungal fermentation product with residual fibrolytic enzymatic activity (MAXFIBER®, Provita Supplements, Pinneberg, Germany) on digestibility, feeding behavior, productive performance, milk quality, and various metabolic parameters of high-producing dairy cows. Thirty-six multiparous Holstein dairy cows were divided into two groups: one group with supplementation of 10 g enzyme/cow/day (GS; n = 18), top-dressed on a total mixed ration, and a control group (CG; n = 18) without supplementation. The results were analyzed using a randomized block design over a 23-day experimental period. Milk yield was evaluated daily, and milk and blood samples were collected on days 0, 14, 19, and 23 to analyze the composition and quality of the milk and blood metabolites. DMI and eating behavior were assessed daily using electronic automatic feeders. Cows that received the enzyme-supplemented diet had higher milk production (p = 0.05), lower somatic cell count (SCC) in milk (p = 0.003), and reduced milk fat (p = 0.01). The assessment of feeding behavior showed a higher frequency of access to feeders and meals per day (p = 0.001) in cows that received enzymes in their diet. Blood glucose levels tended to increase in the group with supplementation of the enzyme product (p = 0.09). However, the treatment had no effect (p > 0.05) on DMI and diet digestibility. Fibrolytic enzymes are a useful tool in dairy cow nutrition, increasing milk yield and reducing SCC in milk. Moreover, the enzymes change feeding behavior by increasing feeding frequency.

Effect of Seasons on Expressions of Milking Parlour Behaviours and Milk Yield of Crossbred Jersey Cows in Tropical Climate

Background: Milking behaviour of dairy cows has remarkable influences on their production efficacy. A number of genetic and environmental features controls and affects lactating behaviour of dairy cow. The aim of present study was to investigate the influence of seasons on milking behaviour and milk production of Jersey crossbred cows in tropical climate. Methods: The experiment was conducted on 94 Jersey crossbred cows maintained at organized dairy farm at ICAR-National Dairy Research Institute, Eastern Regional Station, Kalyani, West Bengal, India and there lactating behavioural features were studied during milking time inside the milking parlour under different seasons. The study period was divided into 3 season’s viz. rainy (September-October), winter (November-February) and summer (March-May). Result: Seasons showed influence upon milking behaviour and milk yield of cows. Temperament of cows were least (1.91±0.03) during winter seasons as compare to rainy and summer (2.15±0.03; 2.09±0.03) respectively and it differed significantly (P less than 0.01), simultaneously daily milk yield were also more (9.36±0.11) kgs during winter seasons as compare to rainy and summer (6.63±0.13; 8.58±0.13) respectively and it differed significantly (P less than 0.01). It can be concluded from the study that seasons showed significant influence on the milking behavioural features and milk yield in Jersey crossbred cows in tropical climate, which suggests that if dairy animals get favourable micro-climatic environment and reduced level of environmental stress they will behave more favourably and simultaneously their production efficacy also inclines.

239 Relationship Between Body Temperature and Behavior-Activities of Non-Pregnant Early Lactation Dairy Cows

Abstract Animal behavior is a component of animal welfare and known to affect physiology as a primary input to the circadian system. Our objectives were to determine if non-pregnant, early lactation dairy cows in a free-stall exhibited circadian rhythms of behavior and core body temperature, and if oscillations in daily behaviors and activities of a cow relate to daily oscillations in body temperature or estrus, parity, and days in milk (DIM). Thirty Holstein cows (21-80 DIM) were observed for a 48-h period. The activities of standing, eating, ruminating, lying, and estrus behavior were recorded every 10 min. Data loggers (ibuttons) secured to blank (progesterone removed) controlled internal drug release (CIDR) devices were placed intravaginally in a subset of cows (n = 11) to record body temperature every 10 min over the same 48 h period. Cows were milked twice each day starting approximately 05:00 and 17:00, and the 70 min period encompassed moving cows to and from the milking parlor. The entire period cows were away from the free-stall was marked as milking. An ethogram was created using a binary system (0,1), with 1 for each behavior. Behavior and temperature data were graphed for visualization, and cosine fit analysis found lack of evidence for circadian rhythms. Programs in R were used to plot behavior to overlay temperature recordings over the 48 h period to determine if temperature varied with behavior-activity. Temperature increased during milking periods in 8 of 11 cows (statistical method in development), but no other relationships were found. To study patterns of behavior, an adjacency matrix was utilized to create a directed graph for each cow. Analysis of the effect of estrus, parity and DIM on behavior patterns is ongoing. Computational approaches are currently being developed that may serve as tools to analyze the effect of production management systems on cow behavior.

Locomotion behavior of dairy cows on traditional summer mountain farms in comparison with modern cubiclehousing without access to pasture.

Pasture based systems enable cattle to express their natural behavior and are thus expected to provide better welfare than the majority of confinement systems. The aim of this study was to objectively measure locomotion activity of healthy dairy cattle kept on mountain pastures (n = 44) compared with cows kept in cubicle housing systems (n = 38). Selected cows were equipped with a validated 3D-accelerometer on one hind limb, and locomotion behavior was recorded for 48 hours. The 1-hour summaries of the variables lying time, walking time, standing bouts, walking bouts and number of strides were summed up to 24-hour summaries, and the means of the stride distance and stride duration were weighted by the variable number of strides. Mountain pasture cows had higher locomotor activity levels in comparison to cubicle cows. Mountain pasture cows spent less time lying down (528.1±109.5 min/day vs. 693.3±73.8 min/day; P<0.0001) and more time walking (75.6±25.9 min/day vs. 38.8±15.8 min/day; P <0.0001) than cubicle cows. Lying bout duration was longer in cubicle than in mountain pasture cows (90.9± 15.2 min/bout vs. 74.2 ± 21.1 min/bout; P = 0.0001), whilst the number of walking bouts was higher in mountain pasture cows than cubicle cows (199.1 ± 49.1 vs. 123.8 ± 43.8 bouts per day; P < 0.001). Likewise, the number of strides was higher in mountain pasture cows than cubicle cows (2040.5 ± 825.3 vs. 916.7 ± 408.6; P < 0.001). Mountain pasture cows had shorter stride duration (P < 0.0001) and shorter strides (P = 0.0002) than cubicle cows (1.8 ± 0.1 s/stride vs 2 ± 0.2 s/stride and 126.3 ± 18.1 vs 142.1 ± 17.8 m/stride, respectively). In summary, cows kept on mountain pasture were more active and spent longer than 12 hours / day standing. Lying markedly less than 12 hours per day seems to represent the normal behavior of pastured cows searching for fresh grass. This does not cause any obvious damage to the locomotor system as claws of cattle are well adapted to long periods of movement on mountain pastures.

A randomised controlled trial to evaluate the impact of indoor living space on dairy cow production, reproduction and behaviour

As a global society, we have a duty to provide suitable care and conditions for farmed livestock to protect animal welfare and ensure the sustainability of our food supply. The suitability and biological impacts of housing conditions for intensively farmed animals is a complex and emotive subject, yet poorly researched, meaning quantitative evidence to inform policy and legislation is lacking. Most dairy cows globally are housed for some duration during the year, largely when climatic conditions are unfavourable. However, the impact on biology, productivity and welfare of even the most basic housing requirement, the quantity of living space, remains unknown. We conducted a long-term (1-year), randomised controlled trial (CONSORT 10 guidelines) to investigate the impact of increased living space (6.5 m2 vs 3 m2 per animal) on critical aspects of cow biology, behaviour and productivity. Adult Holstein dairy cows (n = 150) were continuously and randomly allocated to a high or control living space group with all other aspects of housing remaining identical between groups. Compared to cows in the control living space group, cows with increased space produced more milk per 305d lactation (primiparous: 12,235 L vs 11,592 L, P < 0.01; multiparous: 14,746 L vs 14,644 L, P < 0.01) but took longer to become pregnant after calving (primiparous: 155 d vs 83 d, P = 0.025; multiparous: 133 d vs 109 d). In terms of behaviour, cows with more living space spent significantly more time in lying areas (65 min/d difference; high space group: 12.43 h/day, 95% CI = 11.70–13.29; control space group: 11.42 h/day, 95% CI = 10.73–12.12) and significantly less time in passageways (64 min/d), suggesting enhanced welfare when more space was provided. A key physiological difference between groups was that cows with more space spent longer ruminating each day. This is the first long term study in dairy cows to demonstrate that increased living space results in meaningful benefits in terms of productivity and behaviour and suggests that the interplay between farmed animals and their housed environment plays an important role in the concepts of welfare and sustainability of dairy farming.

Smart Dairy Cattle Farming and In-Heat Detection through the Internet of Things (IoT)

Smart farming is the practice of intelligent agricultural management based on technological data gathered from farm practice for the purpose of increased levels of quality, production and environmental protection. The Internet of Things (IoT) technology is revolutionized in various aspects of agriculture around the world and its application found success in some countries. In this paper, a non-invasive and non-contact estrus detection system integrated IoT technology to improve the detection efficiency of standing-heat behaviors of cows is proposed. The outcome of this study will improve the farm’s management practices through the integration of IoT technology that can remotely monitor the activities of the cows and provide data for analysis and evaluation, will increase the likelihood of future estrus instances for every correct prediction, and will improve the fertility and milk production rates of the cows. Such benefits can contribute to the growth and development of the dairy cattle industry in the Philippines. This study also shows the application of IoT in improving the detection efficiency of standing-heat behaviors of cows through automated detection using Pan-tilt-zoom (PTZ) cameras and a Python-driven Web Application. The dimensions of the barn are measured, and the Cameras' Field of Views (FOVs) is pre-calculated for the strategic positions of the cameras atop of the cowshed. The program detects the cows and any estrus events through the surveillance cameras. The result is sent to the cloud server to display on the web application for analysis. The web app allows updates on cow information, inseminations, pregnancy, and calving records, estimate travel time from the user's geolocation to the farm, provide live monitoring and remote camera accessibility and control through the cameras and deliver reliable cross-platform push-notification and call alerts on the user's device(s) whenever an estrus event is detected. The system initially and correctly detected 4 standing-heat signs, but with 2 false predictions and identifications leading to 2 “True Positive” and 2 “False Positive” results, attaining a 50% detection efficiency. Based on the results, the program performed satisfactorily at 50% detection efficiency.

Assessment of feeding, ruminating and locomotion behaviors in dairy cows around calving - a retrospective clinical study to early detect spontaneous disease appearance.

The study aims to verify the usefulness of new intervals-based algorithms for clinical interpretation of animal behavior in dairy cows around calving period. Thirteen activities associated with feeding-ruminating-locomotion-behaviors of 42 adult Holstein-Friesian cows were continuously monitored for the week (wk) -2, wk -1 and wk +1 relative to calving (overall 30’340 min/animal). Soon after, animals were retrospectively assigned to group-S (at least one spontaneous diseases; n = 24) and group-H (healthy; n = 18). The average activities performed by the groups, recorded by RumiWatch® halter and pedometer, were compared at the different weekly intervals. The average activities on the day of clinical diagnosis (dd0), as well as one (dd-1) and two days before (dd-2) were also assessed. Differences of dd0 vs. dd-1 (ΔD1), dd0 vs. wk -1 (ΔD2), and wk +1 vs. wk -1 (Δweeks) were calculated. Variables showing significant differences between the groups were used for a univariate logistic regression, a receiver operating characteristic analysis, and a multivariate logistic regression model. At wk +1 and dd0, eating- and ruminating-time, eating- and ruminate-chews and ruminating boluses were significantly lower in group-S as compared to group-H, while other activity time was higher. For ΔD2 and Δweeks, the differences of eating- and ruminating-time, as well as of eating-and ruminate-chews were significantly lower in group-S as compared to group-H. Concerning the locomotion behaviors, the lying time was significantly higher in group-S vs. group-H at wk +1 and dd-2. The number of strides was significantly lower in group-S compared to group-H at wk +1. The model including eating-chews, ruminate-chews and other activity time reached the highest accuracy in detecting sick cows in wk +1 (area under the curve: 81%; sensitivity: 73.7%; specificity: 82.4%). Some of the new algorithms for the clinical interpretation of cow behaviour as described in this study may contribute to monitoring animals’ health around calving.

Effects of changes in diet energy density and milking frequency and a single injection of cabergoline at dry-off on feeding behavior and rumination time in dairy cows

Dry-off is a typical management practice, but research on the effects of dry-off on feeding behavior in high-yielding cows is limited. The present study investigated the effects of 2 diet energy densities: lactation diet (normal energy density, NORM) versus a lactation diet diluted with 30% straw (reduced energy density, REDU), both offered ad libitum, and 2 daily milking frequencies (2× vs. 1×) during the 7 d before dry-off day (d 0), and the effects of an injection of either a dopamine agonist [cabergoline (CAB); Velactis, Ceva Santé Animale; labeled for use only with abrupt dry-off; i.e., no reduction in feeding level or milking frequency before the last milking] or saline (SAL) following the last milking on d 0 (2 × 2 × 2 factorial arrangement), on automatically monitored feeding behavior and rumination time in 119 clinically healthy, loose-housed, pregnant, lactating Holstein cows during the week before and after d 0. From d 0, all cows were fed the same dry-cow diet ad libitum. Data were analyzed in R using mixed-effects models. Over the days before d 0, REDU cows spent 30% more time feeding at a 50% lower feeding rate, visited both assigned and unassigned feed bins more frequently, and spent more time ruminating than NORM cows. No clear behavioral effects of reduced milking frequency were found. Within 24 h following injection, CAB cows spent approximately 40% less time feeding at a lower feeding rate, visited their feed bin 28% less often, and spent 40% less time ruminating than SAL cows, irrespective of treatment before dry-off. The current study demonstrates that reducing diet energy density for 1 wk before dry-off led to clear behavioral changes in high-yielding cows. Administering CAB after the last milking induced decreased feeding behavior lasting approximately 24 h, indicating collateral effects other than reduced prolactin secretion.

Basic motion behavior recognition of single dairy cow based on improved Rexnet 3D network

• Rexnet was extended to Rexnet 3D for basic motion behavior recognition of dairy cows. • Sliding window mechanism were used to verify the effectiveness of untrimmed videos. • Rexnet 3D could be used for recognizing the basic motion behavior of dairy cows. Basic motion behaviors of dairy cows (lying, standing and walking) contain abundant health information. However, to monitor these behaviors accurately and efficiently by using camera-based methods remains challenging. In this study, time dimension was added to extend the Rexnet to Rank eXpansion Network 3D algorithm (Rexnet 3D) network to realize non-contact and automatic recognition of dairy cows’ basic motion behaviors. A total of 406 videos containing 256,500 frames of dairy cows in different scenes and postures were collected and tested. Six different networks including Resnet101, Mobilev2, Mobilev3, Shufflev2, C3D and S3D were selected to validate the performance of the proposed network. The average number of frames recognized per second (ANFR), FLOPs and Networks Weight were used to evaluate the performance of the method. Experimental results showed that behavior recognition accuracy, ANFR, FLOPs and Networks Weight of the proposed method were 95.00%, 76.52 fps, 15.80 G and 14.30 M, respectively. Meanwhile, a total of 10 unedited videos with 1,722 s in total (43,065 frames) were collected to verify the effectiveness in real environment. The results showed that the accuracy of the proposed method was higher when the sampling interval was five frames and the accuracy of the specific three basic motion behaviors was 91.02%, and the ANFR was 101.02 fps. In general, these findings indicated that it was feasible to use a designed network for accurate recognition of cows’ basic motion behaviors. The network could provide necessary reference for improving the precision breeding and health welfare of dairy cows.

Cow identification in free-stall barns based on an improved Mask R-CNN and an SVM

• A method for individual identification of cows in unconstrained barn was developed. • The extracted and selected features have strong adaptability to pattern deformation. • A method involving fusion of Mask R-CNN and SVM was used to identify cows. • To reduce the running time and number of model parameters, Mask R-CNN was modified. The identification of individual dairy cows is an important prerequisite for dairy cow behaviour analysis and disease detection. Computer vision-based cow recognition is a noncontact and stress-free approach. In a free environment in a barn, due to changes in camera position and angle, recorded cow patterns are often deformed, making individual cow identification difficult. For cows in an unconstrained barn environment, this paper proposes a method for individual cow identification. First, a top-view image of a cow is obtained, and an improved Mask R-CNN is used to segment this image and extract the shape features of the cow’s back. Then, a Fisher approach is used to select the best feature subset, and a support vector machine (SVM) classifier is applied to identify individual cows. To verify the effectiveness of the target detection algorithm, the proposed method is compared with the traditional Mask R-CNN model, and the precision, recall, F1 score, average run time per image and average precision of the improved Mask R-CNN model are 98.21%, 96.48%, 97.34%, 1.02 s, and 97.39%, respectively. An SVM classifier trained based on the obtained shape features is used for individual cow identification. The proposed method achieves a 98.67% cow identification accuracy based on a dataset containing top-view images of 48 cows. The results demonstrate the effectiveness of the proposed cow identification method and its significant potential for use in precision dairy cow management.

Brief Research Report: How Do Claw Disorders Affect Activity, Body Weight, and Milk Yield of Multiparous Holstein Dairy Cows?

Claw disorders are among the most relevant health problems in dairy herds. Despite being often not clearly visible and not easily detectable for farmers, they may appear as peculiar cow behavioral and performance patterns. This retrospective study aimed to assess cow's behavior and production variations associated with claw disorders. The study involved 54 lactating Italian Holstein cows reared on the same dairy farm. A veterinarian performed the routine hoof trimming every 6 months, diagnosing specific claw disorders. Multiparous cows with no disorders at the first trimming were selected and monitored for the two following trimming sessions. Data coming from the automatic milking system and neck collars and related to the 15 days before a given cow was diagnosed with claw problems during trimming were further collected. These data were compared with those recorded for the same animal over the 15 days preceding the previous trimming in which no claw disorders were observed. Compared to when they had no disorders, the cows affected by claw disorders had a lower daily activity (405 vs. 429 ± 27.7 units/day, p < 0.001), showing a constant decrease in the last 10 days before the trimming, a lower milk yield (26.5 vs. 28.4 ± 1.57 kg/day, p = 0.03), and only a decreasing trend of rumination time. These patterns of activity, milk yield, and rumination characterizing cows affected by claw disorders should promote the development of specific algorithms that would enable early detection of lameness thanks to the deviations of these parameters that are sensitive to cow claw health.