Livestock Welfare Research Articles

Holstein Cattle Face Re-Identification Unifying Global and Part Feature Deep Network with Attention Mechanism.

Simple SummaryAutomated and non-intrusive recognition of individual livestock such as a cow or a pig is a vital requirement for obtaining individual information, something which has played a significant role in the intelligent management and welfare of livestock. However, individual cow identification remains a demanding task mainly due to the quite yet subtle inter-class differences in an open herd setting. The objective of this study is to develop a novel unified global and part feature deep network model (GPN) to learn more discriminative and robust features that can facilitate cow face representation for re-identification and verification. The results of various contrast experiments on collecting the large-scale cow face dataset show that the proposed GPN model outperforms the existing representative re-identification methods, and the further improved GPN-ST model has a higher accuracy rate (up by 2.8% and 2.2% respectively) in Rank-1 and mAP, compared with the GPN model. In conclusion, using the proposed framework can effectively ameliorate the performance of cow face re-identification.In precision dairy farming, computer vision-based approaches have been widely employed to monitor the cattle conditions (e.g., the physical, physiology, health and welfare). To this end, the accurate and effective identification of individual cow is a prerequisite. In this paper, a deep learning re-identification network model, Global and Part Network (GPN), is proposed to identify individual cow face. The GPN model, with ResNet50 as backbone network to generate a pooling of feature maps, builds three branch modules (Middle branch, Global branch and Part branch) to learn more discriminative and robust feature representation from the maps. Specifically, the Middle branch and the Global branch separately extract the global features of middle dimension and high dimension from the maps, and the Part branch extracts the local features in the unified block, all of which are integrated to act as the feature representation for cow face re-identification. By performing such strategies, the GPN model not only extracts the discriminative global and local features, but also learns the subtle differences among different cow faces. To further improve the performance of the proposed framework, a Global and Part Network with Spatial Transform (GPN-ST) model is also developed to incorporate an attention mechanism module in the Part branch. Additionally, to test the efficiency of the proposed approach, a large-scale cow face dataset is constructed, which contains 130,000 images with 3000 cows under different conditions (e.g., occlusion, change of viewpoints and illumination, blur, and background clutters). The results of various contrast experiments show that the GPN outperforms the representative re-identification methods, and the improved GPN-ST model has a higher accuracy rate (up by 2.8% and 2.2% respectively) in Rank-1 and mAP, compared with the GPN model. In conclusion, using the Global and Part feature deep network with attention mechanism can effectively ameliorate the efficiency of cow face re-identification.

Affective State Recognition in Livestock-Artificial Intelligence Approaches.

Simple SummaryEmotions or affective states recognition in farm animals is an underexplored research domain. Despite significant advances in animal welfare research, animal affective state computing through the development and application of devices and platforms that can not only recognize but interpret and process the emotions, are in a nascent stage. The analysis and measurement of unique behavioural, physical, and biological characteristics offered by biometric sensor technologies and the affiliated complex and large data sets, opens the pathway for novel and realistic identification of individual animals amongst a herd or a flock. By capitalizing on the immense potential of biometric sensors, artificial intelligence enabled big data methods offer substantial advancement of animal welfare standards and meet the urgent needs of caretakers to respond effectively to maintain the wellbeing of their animals.Farm animals, numbering over 70 billion worldwide, are increasingly managed in large-scale, intensive farms. With both public awareness and scientific evidence growing that farm animals experience suffering, as well as affective states such as fear, frustration and distress, there is an urgent need to develop efficient and accurate methods for monitoring their welfare. At present, there are not scientifically validated ‘benchmarks’ for quantifying transient emotional (affective) states in farm animals, and no established measures of good welfare, only indicators of poor welfare, such as injury, pain and fear. Conventional approaches to monitoring livestock welfare are time-consuming, interrupt farming processes and involve subjective judgments. Biometric sensor data enabled by artificial intelligence is an emerging smart solution to unobtrusively monitoring livestock, but its potential for quantifying affective states and ground-breaking solutions in their application are yet to be realized. This review provides innovative methods for collecting big data on farm animal emotions, which can be used to train artificial intelligence models to classify, quantify and predict affective states in individual pigs and cows. Extending this to the group level, social network analysis can be applied to model emotional dynamics and contagion among animals. Finally, ‘digital twins’ of animals capable of simulating and predicting their affective states and behaviour in real time are a near-term possibility.

Kesejahteraan Ayam Petelur di PT Inti Tani Satwa Kabupaten Maros

Animal welfare relates to physical and mental states that conform to the natural behavior of animals and need to be implemented to protect animals. Laying hens are very vulnerable to violating animal welfare principles because of high production targets but demands to minimize costs. The study was conducted in the laying hens of the grower phase of PT. ITS with direct observations, interviews, and discussions. The results were analyzed descriptively according to five principles of animal welfare, namely freedom from pain, injury, and disease, freedom from hunger and thirst, freedom from heat and discomfort, freedom from fear and suffering, and freedom to express normal and natural behavior. The results show that the implementation of animal welfare at PT. ITS has been carried out, but there are also aspects that are not fulfilled, namely the aspect of being free to express normal and natural behavior that has not been fully implemented because the density of the rang or cage makes the animals very crowded so this must be corrected so that the livestock welfare aspects can be met.

A scoping review of freight rail noise and vibration impacts on domestic animal health and welfare

AbstractMultiple benefits of freight rail activity have been shown for commercial agribusiness, yet the effects of freight rail-related noise and vibration on domestic livestock health and welfare has so far received little research attention. This scoping review examines peer-reviewed and grey literature addressing associations between freight rail noise, vibration and impacts on domestic livestock. Six databases (Scopus, Science Direct, SAGE, TRID, SPARK, ARRB) were searched for relevant literature published from 1980–2019.PRISMAsearch procedures were used to identify 28 publications relevant to domestic livestock, as well as noise or vibration impact of rail applicable to the freight rail context. Included publications addressed a range of livestock and related species, covering descriptive, review, and experimental findings on noise and vibration impacts. Five publications addressed vibration effects, and 23 addressed noise effects. Effects of noise and vibration on different species indicated that adverse effects vary depending on exposure intensity. The literature indicates that specific thresholds for noise and vibration exposure should be considered when managing freight rail impacts on commercial agribusiness involving avian and mammalian species. Freight rail noise and vibration likely exceeds thresholds for discomfort and harm for avian and mammalian species. Future research should consider case studies that specifically focus on integrating freight rail noise and vibration data to derive species-specific guides for animal health and welfare purposes.

Edible Insects as Food-Insect Welfare and Ethical Aspects from a Consumer Perspective.

Simple SummaryFor prevalent livestock, animal welfare is important to consumers. With increasing interest in edible insects, one might wonder how this concern translates to consumers’ perceptions of the welfare of insects. Therefore, we focus on consumers’ acceptance of how edible insects are currently produced. We first define what animal welfare means for prevalent livestock and transfer relevant aspects to the welfare of insects. Then, we review relevant aspects that shape consumers’ understanding of animal welfare. We provide an overview of the few consumer studies on insect welfare. Last, we present the public discourse on insects and discuss how this might be relevant to consumers’ perceptions of insect welfare.A growing number of studies underline consumers’ concerns about the importance of animal welfare as a general concept for consumers’ purchase decisions. In particular, consumers perceive animal husbandry to be one of the most important aspects of animal welfare. Since intensive livestock production is criticized across society, the acceptance of current intensive production systems of edible insects is an issue of investigation. Criteria of insect welfare might differ from vertebrate welfare. One might argue that it is difficult to define standards for insect welfare due to their large diversity in living environments and feed requirements. In addition, it is debated whether insects are conscious and suffer from pain. It has been demanded to rear insects preferably under natural living conditions and some researchers proposed to consider them as sentient beings. Basic welfare and ethical aspects of insects as food and feed include species-specific mass rearing conditions and euthanasia, i.e., killing procedures. Consumers’ opinions and concerns regarding this issue have hardly been considered so far. In this paper, the animal welfare of prevalent livestock is defined and outlined, and relevant criteria are transferred to insect welfare. Different ways consumers might arrive at an animal welfare understanding are discussed, along with an overview of the few consumer studies on insect welfare. Furthermore, we consider how insects are presented in the public discourse and infer how this might be relevant to consumers’ perceptions of insect welfare.

CattleFaceNet: A cattle face identification approach based on RetinaFace and ArcFace loss

Cattle identification is crucial to be registered for breeding association, food quality tracing, disease prevention and control and fake insurance claims. Traditional non-biometrics methods for cattle identification is not really satisfactory in providing reliability due to theft, fraud, and duplication. In this study, a computer vision technique was proposed to facilitate precision animal management and improve livestock welfare. This paper presents a novel face identification framework by integrating light-weight RetinaFace-mobilenet with Additive Angular Margin Loss (ArcFace), namely CattleFaceNet. RetinaFace-mobilenet is designed for face detection and location, and ArcFace is adopted to strengthen the within-class compactness and also between-class discrepancy during training. Experiments on real-word scenarios dataset prove that RetinaFace-mobilenet achieves superior detection performance and significantly accelerates the computation time against RetinaNet. Three loss functions utilized in human face recognition combined with RetinaFace-mobilenet are compared and results indict that the proposed CattleFaceNet outperforms others with identification accuracy of 91.3% and processing time of 24 frames per second (FPS). This research work demonstrates the potential candidate of CattleFaceNet for livestock identification in real time in practical production scenarios.

Early Phenotype Programming in Birds by Temperature and Nutrition: A Mini-Review

Early development is a critical period during which environmental influences can have a significant impact on the health, welfare, robustness and performance of livestock. In oviparous vertebrates, such as birds, embryonic development takes place entirely in the egg. This allows the effects of environmental cues to be studied directly on the developing embryo. Interestingly, beneficial effects have been identified in several studies, leading to innovative procedures to improve the phenotype of the animals in the long term. In this review, we discuss the effects of early temperature and dietary programming strategies that both show promising results, as well as their potential transgenerational effects. The timing, duration and intensity of these procedures are critical to ensure that they produce beneficial effects without affecting animal survival or final product quality. For example, cyclic increases in egg incubation temperature have been shown to improve temperature tolerance and promote muscular growth in chickens or fatty liver production in mule ducks. In ovo feeding has also been successfully used to enhance digestive tract maturation, optimize chick development and growth, and thus obtain higher quality chicks. In addition, changes in the nutritional availability of methyl donors, for example, was shown to influence offspring phenotype. The molecular mechanisms behind early phenotype programming are still under investigation and are probably epigenetic in nature as shown by recent work in chickens.

Radio frequency-based periodic cloud data analysis for smart farming

Radio frequency-based smart farming is revolutionising agriculture. To meet the growing demand of food, farm productivity should be improved by predicting crop performance in diverse environmental conditions. Farming becomes smarter with automation and radio frequency technology. Radio frequency collect and manage data about crop performance, climate change, livestock welfare, resource shortages and atmosphere that frequently influence farming construction via smartphone applications where the data container remains and is retrieved wherever. It also aids in animal nursing and termination. With smart sensors and equipment, farmers can increase crop production with reduced cost and time. Smart farming along with sensors offers the guarantee of expanding efficiency while diminishing generation costs and limiting ecological effects. The sensor data will be collected and stored in the cloud. Analysis of farming is done with computing the data stored in the cloud. Based on these cloud data, farmers perform farming in a proper annual time. Cloud-based storage is more appropriate for these computations. Various climatic conditions for farming are analysed with these cloud data.

The role of animal welfare in improving the future of farming

Context Animal production plays a critical role in many global challenges around sustainability, including climate change and resilience, One Health and food security. With this role comes pressures on livestock welfare. Aims This paper demonstrates key contributions animal welfare makes to global sustainability challenges. Methods This paper highlights ‘win–win’ improvements for both animal welfare and other aspects of sustainability by using the following four case studies: tail docking Australian sheep, agroforestry systems in Ethiopia, the Australian dairy-beef industry, and strategic feeding of goats in Pakistan. Key results These case studies show how animal welfare can be improved alongside livelihoods. However, even in these win–win situations, the adoption of improved practices is not guaranteed. Conclusions Long-term, sustained change in animal welfare can simultaneously make in roads to other challenges around sustainability. To do this, we must have a broader understanding of the system in which the animals are raised, so that barriers to change can be identified. Implications Lessons from these case studies can be applied to other production contexts and challenges, highlighting the universal value of understanding and addressing animal welfare.

A pioneer in preslaughter animal welfare

Carmen Gallo Stegmaier spent her working life in academia in Chile. She recently won a UK Humane Slaughter Association award for her dedication to enhancing livestock welfare in Latin America.

Metabolic Potential of Epichloë Endophytes for Host Grass Fungal Disease Resistance.

Asexual species of the genus Epichloë (Clavicipitaceae, Ascomycota) form endosymbiotic associations with Pooidae grasses. This association is important both ecologically and to the pasture and turf industries, as the endophytic fungi confer a multitude of benefits to their host plant that improve competitive ability and performance such as growth promotion, abiotic stress tolerance, pest deterrence and increased host disease resistance. Biotic stress tolerance conferred by the production of bioprotective metabolites has a critical role in an industry context. While the known antimammalian and insecticidal toxins are well characterized due to their impact on livestock welfare, antimicrobial metabolites are less studied. Both pasture and turf grasses are challenged by many phytopathogenic diseases that result in significant economic losses and impact livestock health. Further investigations of Epichloë endophytes as natural biocontrol agents can be conducted on strains that are safe for animals. With the additional benefits of possessing host disease resistance, these strains would increase their commercial importance. Field reports have indicated that pasture grasses associated with Epichloë endophytes are superior in resisting fungal pathogens. However, only a few antifungal compounds have been identified and chemically characterized, and these from sexual (pathogenic) Epichloë species, rather than those utilized to enhance performance in turf and pasture industries. This review provides insight into the various strategies reported in identifying antifungal activity from Epichloë endophytes and, where described, the associated antifungal metabolites responsible for the activity.

Strengthening livestock welfare policies to mitigate human–wildlife conflicts

Strengthening livestock welfare policies to mitigate human–wildlife conflicts

The Markers of Stress in Swine Oral Fluid.

IntroductionThe study measured the hormonal and protein markers of acute stress, those of oxidative stress and total antioxidant capacity (TAC) in swine oral fluid, determined which of these parameters would be the most appropriate for future livestock welfare assessment and established the time when the samples should be taken.Material and MethodsStress was induced in 7 out of 14 castrated six-week-old Danbred×Duroc pigs by immobilisation on a nasal snare at 8 a.m., 1 p.m., and 6 p.m. and samples were taken both directly after the stressor was applied and 30 min later. The remaining pigs were the control group, which were not immobilised; their samples were taken at the same times. The concentrations of hormones and malondialdehyde (MDA) were measured using liquid chromatography with tandem mass spectrometry, while those of alpha-amylase and TAC were measured using spectrophotometry.ResultsThe levels of cortisol and cortisone increased with statistical significance immediately after the acute stress response and 30 min later. A cut-off value set at 0.25 ng/mL cortisol concentration was capable of distinguishing between the stressed and control groups with 100% accuracy in evening samples and 95% accuracy overall. Prednisolone was not present, and the levels of testosterone and corticosterone were low and not distinctive. Alpha-amylase became significantly more concentrated during stress induction and 30 min later. The TAC and MDA levels rose after the stress but without statistical significance.ConclusionThe most suitable markers of acute stress were cortisol, cortisone and alpha-amylase. Oral fluid is a reliable material for monitoring the level of pigs’ stress and should be collected in the evening.

Automated Muzzle Detection and Biometric Identification via Few-Shot Deep Transfer Learning of Mixed Breed Cattle

Livestock welfare and management could be greatly enhanced by the replacement of branding or ear tagging with less invasive visual biometric identification methods. Biometric identification of cattle from muzzle patterns has previously indicated promising results. Significant barriers exist in the translation of these initial findings into a practical precision livestock monitoring system, which can be deployed at scale for large herds. The objective of this study was to investigate and address key limitations to the autonomous biometric identification of cattle. The contributions of this work are fourfold: (1) provision of a large publicly-available dataset of cattle face images (300 individual cattle) to facilitate further research in this field, (2) development of a two-stage YOLOv3-ResNet50 algorithm that first detects and extracts the cattle muzzle region in images and then applies deep transfer learning for biometric identification, (3) evaluation of model performance across a range of cattle breeds, and (4) utilizing few-shot learning (five images per individual) to greatly reduce both the data collection requirements and duration of model training. Results indicated excellent model performance. Muzzle detection accuracy was 99.13% (1024 × 1024 image resolution) and biometric identification achieved 99.11% testing accuracy. Overall, the two-stage YOLOv3-ResNet50 algorithm proposed has substantial potential to form the foundation of a highly accurate automated cattle biometric identification system, which is applicable in livestock farming systems. The obtained results indicate that utilizing livestock biometric monitoring in an advanced manner for resource management at multiple scales of production is possible for future agriculture decision support systems, including providing useful information to forecast acceptable stocking rates of pastures.

COVID-19 Syndrome: Nexus with Herbivory and Exposure Dynamics for Monitoring Livestock Welfare and Agro-Environment

The outbreak of severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) is a public health emergency that turns the year 2020–2021 into annus horribilis for millions of people across international boundaries. The interspecies transmission of this zoonotic virus and mutated variants are aided by exposure dynamics of infected aerosols, fomites and intermediate reservoirs. The spike in the first, second and third waves of coronavirus confirms that herd immunity is not yet reached and everyone including livestock is still vulnerable to the infection. Of serious concern are the communitarian nature of agrarians in the livestock sector, aerogenous spread of the virus and attendant cytocidal effect in permissive cells following activation of pathogen recognition receptors, replication cycles, virulent mutations, seasonal spike in infection rates, flurry of reinfections and excess mortalities that can affect animal welfare and food security. As the capacity to either resist or be susceptible to infection is influenced by numerous factors, identifying coronavirus-associated variants and correlating exposure dynamics with viral aerosols, spirometry indices, comorbidities, susceptible blood types, cellular miRNA binding sites and multisystem inflammatory syndrome remains a challenge where the lethal zoonotic infections are prevalent in the livestock industry, being the hub of dairy, fur, meat and egg production. This review provides insights into the complexity of the disease burden and recommends precision smart-farming models for upscaling biosecurity measures and adoption of digitalised technologies (robotic drones) powered by multiparametric sensors and radio modem systems for real-time tracking of infectious strains in the agro-environment and managing the transition into the new-normal realities in the livestock industry.

The Quest for Genes Involved in Adaptation to Climate Change in Ruminant Livestock.

Simple SummaryAfter domestication in specific regions, livestock followed human migrations and colonized the whole world. During this population expansion, human and natural selection, together with demographic events, molded the livestock genome leading to local breeds and populations able to produce milk, meat, wool and tractive power in many different agro-climatic conditions. The climate is changing, with temperatures and the frequency of extreme climatic events increasing, which affects livestock welfare and production efficiency, particularly of the highly productive breeds. Genomics is now able to explore the DNA of local breeds adapted to extreme environments in search of genes carrying signatures of selection for adaptation. This review summarizes methods used to accomplish this task, giving examples of results achieved and perspectives for future breeding.Livestock radiated out from domestication centres to most regions of the world, gradually adapting to diverse environments, from very hot to sub-zero temperatures and from wet and humid conditions to deserts. The climate is changing; generally global temperature is increasing, although there are also more extreme cold periods, storms, and higher solar radiation. These changes impact livestock welfare and productivity. This review describes advances in the methodology for studying livestock genomes and the impact of the environment on animal production, giving examples of discoveries made. Sequencing livestock genomes has facilitated genome-wide association studies to localize genes controlling many traits, and population genetics has identified genomic regions under selection or introgressed from one breed into another to improve production or facilitate adaptation. Landscape genomics, which combines global positioning and genomics, has identified genomic features that enable animals to adapt to local environments. Combining the advances in genomics and methods for predicting changes in climate is generating an explosion of data which calls for innovations in the way big data sets are treated. Artificial intelligence and machine learning are now being used to study the interactions between the genome and the environment to identify historic effects on the genome and to model future scenarios.

Emotion Recognition in Horses with Convolutional Neural Networks

Creating intelligent systems capable of recognizing emotions is a difficult task, especially when looking at emotions in animals. This paper describes the process of designing a “proof of concept” system to recognize emotions in horses. This system is formed by two elements, a detector and a model. The detector is a fast region-based convolutional neural network that detects horses in an image. The model is a convolutional neural network that predicts the emotions of those horses. These two elements were trained with multiple images of horses until they achieved high accuracy in their tasks. In total, 400 images of horses were collected and labeled to train both the detector and the model while 40 were used to test the system. Once the two components were validated, they were combined into a testable system that would detect equine emotions based on established behavioral ethograms indicating emotional affect through the head, neck, ear, muzzle, and eye position. The system showed an accuracy of 80% on the validation set and 65% on the test set, demonstrating that it is possible to predict emotions in animals using autonomous intelligent systems. Such a system has multiple applications including further studies in the growing field of animal emotions as well as in the veterinary field to determine the physical welfare of horses or other livestock.

Butyric and Citric Acids and Their Salts in Poultry Nutrition: Effects on Gut Health and Intestinal Microbiota.

Intestinal dysfunction of farm animals, such as intestinal inflammation and altered gut microbiota, is the critical problem affecting animal welfare, performance and farm profitability. China has prohibited the use of antibiotics to improve feed efficiency and growth performance for farm animals, including poultry, in 2020. With the advantages of maintaining gut homeostasis, enhancing digestion, and absorption and modulating gut microbiota, organic acids are regarded as promising antibiotic alternatives. Butyric and citric acids as presentative organic acids positively impact growth performance, welfare, and intestinal health of livestock mainly by reducing pathogenic bacteria and maintaining the gastrointestinal tract (GIT) pH. This review summarizes the discovery of butyric acid (BA), citric acid (CA) and their salt forms, molecular structure and properties, metabolism, biological functions and their applications in poultry nutrition. The research findings about BA, CA and their salts on rats, pigs and humans are also briefly reviewed. Therefore, this review will fill the knowledge gaps of the scientific community and may be of great interest for poultry nutritionists, researchers and feed manufacturers about these two weak organic acids and their effects on intestinal health and gut microbiota community, with the hope of providing safe, healthy and nutrient-rich poultry products to consumers.

Exploring farmers' understanding of and responses to endemic animal health and welfare issues in the UK.

This paper uses two endemic health conditions to explore farmer understandings of and responses to livestock health and welfare issues. The findings are based on a survey of 42 livestock farmers in the north of England, exploring how they manage lameness in sheep and cattle and bovine viral diarrhoea in cattle. We identify similarities and differences in their approaches. Two themes emerge. (1) The importance of difference between animal types (i.e., beef cattle, dairy cattle and sheep), which highlights the 'complex' and 'multifactorial' nature of animal health and welfare. It is necessary to unpack this to understand the interplay of animal, resource and management issues in farmer responses. (2) Previous research has identified 'lack of knowledge' as a key welfare issue. Our findings reveal farmers are in fact seeking, acquiring and sharing knowledge on practices related to the management of animal health however individual circumstance and context influence how this translates in practice. Our research highlights the importance of integrating different perspectives and knowledges as a way of understanding and responding to animal health and welfare concerns. Facilitating knowledge exchange both within and between different groups and sectors is vital in achieving this.

Predicting of mutton sheep stress coupled with multi-environment sensing and supervised learning network in the transportation process

Stress response caused by transportation is an important reason that affects the health and welfare of livestock, and the intelligent and precise monitoring of the stress state during transportation is a blank in research. On-site tracking experiments, literature review, and expert consultation indicate that environmental temperature, relative humidity and vibration intensity of vehicle are important stressors in the transportation process, and cortisol (COR) and adrenocorticotropic hormone (ACTH) are the key biomarkers to characterize the stress level. In this paper, stressors causing stress response were collected through environmental sensor network, and key biomarkers representing stress response were identified through biochemical analysis, and their change rules were analyzed and compared. Furthermore, the predictive models coupled with environmental parameters and stress markers were constructed through supervised learning networks. The results showed that: (1) After testing, the data collection performance of the multi-sensor network is reliable (P < 0.05); transportation will cause significant changes in the concentration of COR and ACTH (P < 0.05); the simulation results of stress markers data have no significant difference compared with the sample data (PSpline > PPchip > 0.90). (2) The validation results of models showed that: for COR, the absolute error and relative error based on general regression neural network (GRNN) are 2.59 ± 0.16 ng/mL and 1.30 ± 0.07%, respectively, which is better than the accuracy based on back propagation neural network (BPNN) and Elman neural network (Elman); for ACTH, the absolute error and relative error based on GRNN models were 1.75 ± 0.13 pg/mL and 2.14 ± 0.17%, respectively, which were better than the accuracy based on BPNN and Elman models. (3) For the prediction performance, the prediction accuracy of GRNN model for COR concentration is basically equal to that of Elman model, which is 0.17% higher than that of BPNN model for COR concentration, but the maximum relative error of BPNN model for COR concentration is the smallest. The prediction accuracy of Elman model for ACTH concentration is 0.09% higher than that of GRNN model, and 0.29% higher than that of BPNN model. But the maximum relative error of Elman model for ACTH concentration is the largest, and the running time is significantly longer. Although the performance of the three modeling methods is different, their overall accuracy has reached more than 97%. Therefore, this study can effectively identify the stress state of mutton sheep in the actual transportation process, and provide technical support for the health traceability and control of mutton sheep.