No description available.

Ever since 1976, livestock farmers in Fiji have been intrigued about biogas digesters as an animal waste management measure, but the concept has never evolved into one that is sustainable. Renewed interest came in 1997 after a better Chinese Modified Dome (CMD) design was established, with this came government funding, however by 2002, some failure was also observed. This study set out to investigate the reasons why biogas digesters have never been established as a sustainable concept, its effectiveness as an animal waste management tool and what changes if any, can be made to try and influence the further development of the concept in Fiji. In order to achieve the objectives, which were simply obtaining the answers to the questions presented above, several different approaches had to be followed. This study was the first on biogas digesters in Fiji and it was unique in the sense that the major component dealt with the social aspects of farmers in relation to their biogas digesters. Basically the study was carried out in two parts - desktop studies and field studies. Desktop studies were done to better understand the situation while the field studies included semi-structured interviews with the farmers as well as the collection of biophysical data. Twelve outcomes are presented in chapter 7 of this paper. Two in particular are the improvement of construction and maintenance, which can only come about through training. The research questions are also answered in chapter 7, with recommendations put forward on possible directions to take in terms of trying to influence the development of the concept in Fiji / Master of Science (Hons)

Livestock farmers on communal land rely heavily on state veterinary services (SVS) for animal health care. State provided primary health care services are however, not readily accessible to many such communities. The study was conducted to investigate the primary animal health care (PAHC) delivery methods that are preferred by communal land-based livestock farmers, with special focus on the possibility of using community-based animal health workers (CAHW). The study was conducted in Mnisi community in Bushbuckridge local municipality of Mpumalanga Province. Questionnaire-based interviews were conducted with 133 cattle farmers from 10 of the 15 inspection points in Mnisi community. After analysis of the data from the questionnaire surveys, group discussions were held at five inspection points to obtain clarity on the community’s needs and expectations for the delivery of PAHC services. Data was captured and analysed in Microsoft Excel using descriptive statistics. Responses from the group discussions and other interviews were captured and reported on. The respondents were predominately male (72 % of all respondents), owners of the livestock (80%) and had an average of 13 cattle each. The major reasons given for the exit of cattle from the herds were sales (38% of exits), drought (18.2%), stock theft (15.6%) and diseases (14.8%). Overall 55.1% of exits were losses while 43.7% were for beneficial purposes (1.2% of the exits from the cattle herds were not explained). The farmers perceived nutritional problems, animal diseases, access to water, stock theft, drought and ticks/dipping the six important constraints to livestock production. Amongst diseases and disease conditions, tick burdens, lumpy skin disease and internal parasites were considered the major challenges. Most farmers (77%) did not vaccinate their cattle. The main reason that was given for not vaccinating cattle was that vaccines were provided free by the SVS. None of the farmers felt that the vaccines were not available, not effective or too cumbersome to apply. Despite the free vaccines, 67% did not know if their cattle were ever vaccinated and or what diseases the administered vaccines prevented. All farmers controlled ectoparasite infestation either by plunge dipping (90.2%) or use of pour-on acaricides (9.8%). Furthermore, 78% of those who dipped their cattle also used other methods to control ectoparasites, popularly, whole body and/or spot spraying as and when it was necessary. Endoparasites were controlled by 59.4% of the respondents, predominately by drenching as and when it was necessary. The alternative products for ectoparasites and endoparasites were obtained mainly from the farmers’ cooperative in Hoedspruit. For 99% of the respondents the primary providers for animal health care in Mnisi were SVS professionals (predominately Animal Health Technicians, AHTs). Other community members were consulted by 30% of the respondents. The common routes of contact were visiting the SVS offices, contact at the inspection points and calling the officials with a cell phone. Two-thirds of the farmers reported diseases to SVS and most of them (94%) were happy with the response to the reports. In general the farmers were satisfied with the accessibility of the AHTs (72%), state veterinarians (95%) and the animal clinic staff (77%). They were also satisfied with the information that they received from SVS. The most preferred providers of animal health care were the AHTs (77%). When prompted, 92% of the respondents said they would use the services of a community-based person trained to handle animal health problems, mainly for advice on disease management, drug administration, disease surveillance, diagnosis and castration. During the group discussions, facilitating access to fodder and water were included in the tasks for the community-based persons. Forty-two percent of the respondents said that they would use such a person frequently and 83% were willing to pay for the services in cash and/or kind. The general impression was that young people should be trained to carry out these tasks. Farmers preferred that the community-based person should work closely with the AHTs and be in contact with the farmers as frequently as 2–3 times a week to monthly. It is concluded that farmers of Mnisi community were satisfied with the availability and accessibility of the PAHC services provided by the SVS, and in fact preferred the SVS for this service. The farmers however lacked the knowledge and ability to adequately handle the production, animal health and livestock security problems with which they were confronted. A community-based service would therefore be beneficial if it holistically addressed the critical challenges of access to fodder and water, control of diseases and ticks, ensured livestock security, and also equipped farmers to manage these challenges. It is therefore recommended that Community-based Livestock Workers could be engaged if they are adequately qualified to fulfil this role. The use of Farmer Field Schools facilitated either by AHTs or CLWs for capacitation of the farmers should be investigated. The cost implications, potential benefits and effectiveness of selected options should be explored further before conclusive decisions are taken. It is further recommended that some effort should be invested in stimulating community cohesion amongst farmers to enable them to collectively address challenges that are common to them. The needs for the delivery of PAHC services in Mnisi community illustrate that such services should be tailor-made for the intended beneficiary communities, taking into consideration the existing support structures, the communities’ socio-cultural issues, cost effectiveness and the possible impact of the services. Copyright / Dissertation (MSc)--University of Pretoria, 2012. / Veterinary Tropical Diseases / unrestricted

Precision livestock farming technologies have been widely researched over the last decade. These technologies help in monitoring animal health and welfare parameters in a continuous, automated fashion. Under this umbrella of precision livestock farming, this study focuses on activity classification and body weight prediction in pigs. Activity monitoring is essential for understanding the health and growth of pigs. To automate this task effectively, we propose efficient and accurate sensor-based deep learning (DL) solutions. Among these, the 2D Residual Networks emerged as the best performing model, achieving an accuracy of 95.6%. This accuracy was 15.6% higher than that of other machine learning approaches. Additionally, accurate pig weight estimation is crucial for pork production, as it provides valuable insights into growth rates, disease prevalence, and overall health. Traditional manual methods of estimating pig weights are time-consuming and labor-intensive. To address this issue, we propose a novel approach that utilizes deep learning techniques on depth images for weight prediction. Through a custom image preprocessing pipeline, we train DL models to extract meaningful information from depth images for weight prediction. Our findings show that XceptionNet gives promising results, with a mean absolute error of 2.82 kg and a mean absolute percentage error of 7.42%. In comparison, the best performing statistical model, support vector machine, achieved a mean absolute error of 4.51 kg mean absolute percentage error of 15.56%. / Master of Science / With the increasing demand for food production in recent decades, the livestock farming industry faces significant pressure to modernize its methods. Traditional manual tasks such as activity monitoring and body weight measurement have been time-consuming and labor-intensive. Moreover, manual handling of animals can cause stress, negatively affecting their health. To address these challenges, this study proposes deep learning-based solutions for both activity classification and automated body weight prediction. For activity classification, our solution incorporates strategic data preprocessing techniques. Among various learning techniques, our deep learning model, the 2D Residual Networks, achieved an accuracy of 95.6%, surpassing other approaches by 15.6%. Furthermore, this study also compares statistical models with deep learning models for the body weight prediction task. Our analysis demonstrates that deep learning models outperform statistical models in terms of accuracy and inference time. Specifically, XceptionNet yielded promising results, with a mean absolute error of 2.82 kg and a mean absolute percentage error of 7.42%, outperforming the best statistical model by nearly 8%.

La cité de Nîmes, territoire localisé en bordure de la mer Méditerranée (Gard et partie est de l’Hérault), est créée au cours du Ier s. av. J.‑C., avec à sa tête un chef‑lieu (Nemausus) auquel sont rattachés de nombreux établissements répartis au sein de ce vaste territoire. La société nîmoise, tout en conservant son héritage protohistorique, est profondément influencée par le modèle socio‑économique romain dans ce nouveau cadre administratif centralisé. Dans ce contexte de transition, il était intéressant d’aborder la thématique de l’alimentation carnée des habitants de cette cité entre le IIème s. av. J.‑C. et le IIème s. ap. J.‑C. L’étude réalisée à partir des restes fauniques des mammifères provenant du chef‑lieu, d’agglomérations secondaires et d’établissements ruraux, permet d’éclairer l’exploitation des animaux dits « de bouche » dans cette cité. L’analyse propose d’aborder les questions des zones d’élevage et de la présence du bétail à l’intérieur des villes, la gestion des cheptels qui révèle des différences d’exploitation entre les troupeaux et une diversité des productions : lait, laine, force de travail et bien évidemment la viande. Les données archéozoologiques ont également permis d’observer des différences entre sites et des variations chronologiques dans le choix des espèces consommées, ainsi que dans les techniques de découpe des carcasses, les qualités de viande et la gestion des déchets d’origine animale. Les résultats font apparaître une organisation territoriale de l’économie animale nîmoise avec des lieux de production et de consommation, des activités que l’on retrouve souvent entremêlées au sein des établissements. / The city of Nîmes is located near the Mediterranean Sea (Gard and eastern Hérault provinces). Created during the first century BC, the city consists of several establishments scattered in a vast territory and connected to a chief town (Nemausus). In this centralized administrative organisation, a new society appears which, although retaining its protohistoric heritage, is deeply influenced by the Roman socio‑economic model. In this transitionnal context, the question of the meat diet of the inhabitants of this city between the 2nd century BC and the 2nd century AD is of great interest. This study is based on the faunal remains of mammals founded at several archaeological sites: chief town, secondary agglomerations and some rural settlements. This work will focus on the animals which are eaten. The data analysis raises the question of livestock production areas or at least the presence of animals inside the urban sites. Livestock management reveals differences between herds and a diversity of productions: milk, wool, traction and of course meat. The zooarchaeological data also revealed differences between sites and chronologies concerning the choice of species consumed, butchering systems, meat quality and management of animal waste. All these results suggest a territorial organization of animal economy based on production and consumption areas, although these activities are often mixed inside the establishments.

Les exploitations de polyculture-élevage sont de plus en plus reconnues au plan international comme limitant les problèmes environnementaux tout en permettant une agriculture productive et économiquement viable. Les incitations à la spécialisation de la PAC et la diminution de la main d'oeuvre agricole les ont néanmoins marginalisées en Europe. Cette thèse vise à éclairer les conditions de maintien d'exploitations de polyculture-élevage durables. Un travail en partenariat avec des agriculteurs, maires et conseillers agricoles a été conduit sur un cas d'étude en zone défavorisée simple française, les Coteaux de Gascogne. Nos travaux ont montré que, malgré une forte variabilité entre exploitations, la polyculture-élevage locale était un bon compromis par rapport à la spécialisation pour une durabilité environnementale et économique. Une analyse des trajectoires passées des exploitations nous a permis d'éclairer quatre types de « chemins pour durer » en polyculture-élevage. Parmi ceux-ci, deux sont apparus pertinents pour le maintien de la polyculture-élevage en contexte incertain : « maximiser l'autonomie » et « diversifier les ateliers ». Sur la base de ces deux types de trajectoires, nous avons co-construit avec les partenaires deux scénarios techniques prospectifs. Pour le type « maximiser l'autonomie », implanter des intercultures fourragères basées sur des légumineuses permettrait de favoriser l'autonomie alimentaire du troupeau en maintenant la fertilité des sols. Pour le type « diversifier les ateliers », finir des génisses permettrait de les valoriser en circuit court. Ces scénarios ont été adaptés puis simulés sur des exploitations locales. Ce type de démarche a permis i) d'impliquer fortement les acteurs locaux via des réunions collectives et ii) une approche prospective originale fondée sur une étude rétrospective intégrant le temps long. / Mixed crop-livestock farms are again attracting worldwide interest, as they are considered to be a good way to limit environmental problems while allowing a productive and economically viable agriculture. The incentives of the Common Agricultural Policy and decreasing workforce availability nevertheless marginalized these farms in Europe. This thesis aims at understanding the conditions for a survival of sustainable mixed crop-livestock farms. A partnership process with farmers, mayors and technical advisers has been led in a French less favoured area, the Coteaux de Gascogne. Our work has shown that even if a wide variability existed between farms, local mixed crop-livestock farming was a good trade-off compared to farm specialization concerning an environmental and economic sustainability. An analysis of farm past trajectories allowed us to enlighten four “paths to last” in mixed crop-livestock farming. Two of these ones appeared to be suitable paths to maintain mixed crop-livestock farms in a uncertain context: “maximizing autonomy” and “diversification of production units”. On the basis of these two types of trajectories, we have co-constructed with local actors two technical prospective scenarios. In line with the type “maximizing autonomy”, forage legume intercropping could enable to autonomously feed the bovine herd while maintaining soil fertility. In line with the type “diversification of production units”, fattening heifers would allow a commercialization through short circuit. These scenarios have been adapted and then simulated on local farms. This type of approach allowed to i) strongly involve local actors through collective meetings and ii) an original future study based on a retrospective study integrating long time changes.

A method of continuously monitoring weight would aid producers by ensuring all pigs are gaining weight and increasing the precision of marketing pigs thus saving money. Electronically monitoring weight without moving the pigs to the scale would eliminate a stress-generating source. Therefore, the development of methods for monitoring the physical conditions of animals from a distance appears as a necessity for obtaining data with higher quality. In pigs\' production, animals\' weighing is a practice that represents an important role in the control of the factors that affect the performance of the herd and it is an important factor on the production\'s monitoring. Therefore, this research aimed to extract weight data of pigs through depth images. First, a validation of 5 Kinect &reg; depth sensors was completed to understand the accuracy of the depth sensors. In addition, equations were generated to correct the dimensions\' data (length, area and volume) provided by these sensors for any distance between the sensor and the animals. Depth images and weights of finishing pigs (gilts and barrows) of three commercial lines (Landrace, Duroc and Yorkshire based) were acquired. Then, the images were analyzed with the MATLAB software (2016a). The pigs on the images were selected by depth differences and their volumes were calculated and then adjusted using the correction equation developed. Also, pigs\' dimensions were acquired for updating existing data. Curves of weight versus corrected volumes and corrected dimensions versus weight were adjusted. Equations for weight predictions through volume were adjusted for gilts and barrows and for each of the three commercial lines used. A reduced equation for all the data, without considering differences between sexes and genetic lines was also adjusted and compared with the individual equations using the Efroymson\'s algorithm. The result showed that there was no significant difference between the reduced equation and the individual equations for barrows and gilts (p<0.05), and the global equation was also no different than individual equations for each of the three sire lines (p<0.05). The global equation can predict weights from a depth sensor with an R2 of 0,9905. Therefore, the results of this study show that the depth sensor would be a reasonable approach to continuously monitor weights. / Um método de monitoramento contínuo da massa corporal de suínos auxiliaria os produtores, assegurando que todos os animais estão ganhando massa e aumentando a sua precisão de comercialização, reduzindo-se perdas. Obter eletronicamente a massa corporal sem mover os animais para a balança eliminaria uma fonte geradora de estresse. Portanto, o desenvolvimento de métodos para monitorar as condições físicas dos animais à distância se mostra necessário para a obtenção de dados com maior qualidade. Na produção de suínos, a pesagem dos animais é uma prática que representa um papel importante no controle dos fatores que afetam o desempenho do rebanho e o monitoramento da produção. Portanto, esta pesquisa teve como objetivo extrair, automaticamente, dados de massa de suínos por meio de imagens em profundidade. Foi feita, primeiramente, uma validação de 5 sensores de profundidade Kinect &reg; para compreender seu comportamento. Além disso, foram geradas equações para corrigir os dados de dimensões (comprimento, área e volume) fornecidos por estes sensores para qualquer distância entre o sensor e os animais. Foram obtidas imagens de profundidade e massas corporais de suínos e crescimento e terminação (fêmeas e machos castrados) de três linhagens comerciais (Landrace, Duroc e Yorkshire). Em seguida, as imagens foram analisadas com o software MATLAB (2016a). Os animais nas imagens foram selecionados por diferenças de profundidade e seus volumes foram calculados e depois ajustados utilizando a equação de correção desenvolvida. Foram coletadas, ainda, dimensões dos animais para atualização de dados existentes. Curvas de massa versus volumes corrigidos e de dimensões corrigidas versus massa, foram ajustadas. Equações para predição de massa a partir do volume foram ajustadas para os dois sexos e para as três linhagens comerciais. Uma equação reduzida, sem considerar as diferenças entre sexos e linhagens, também foi ajustada e comparada com as equações individuais utilizando o algoritmo de Efroymson. O resultado mostrou que não houve diferença significativa entre a equação reduzida e as equações individuais tanto para sexo (p <0,05), quanto para linhagens (p <0,05). A equação global pode predizer massas a partir do volume obtido com o sensor, com um R2 de 0,9905. Portanto, os resultados deste estudo mostram que o sensor de profundidade é uma abordagem razoável para monitorar as massas dos animais.

Os mercados da pecuária familiar podem ser entendidos como convenções, visto que essas explicam a coordenação das ações dos diferentes atores, uma vez que contemplam a diversidade de dinâmicas que podem existir nos diferentes mercados. Pode-se dizer que, se não ativamente planejadas e constituídas pelos próprios atores, as relações de mercados são ao menos mediadas por eles, pois enquanto alguns afastam seus processos de produção e trabalho do mercado, outros neles se engajam. Por essa razão, não basta compreender por que determinados atores tomam dadas decisões, é preciso compreender por que o fazem e quais as consequências dessas decisões. Diante disso, o objetivo desta tese é compreender como são constituídos os mercados da pecuária familiar, mediante uma análise das características que orientam as interações entre os atores sociais dos distintos canais de comercialização e da identificação e caracterização dos diferentes tipos de mercados e das razões de sua estruturação. Tendo como objeto de estudo a pecuária familiar, foram realizadas entrevistas com 39 pecuaristas familiares nos municípios de Bagé, Dom Pedrito e Pinheiro Machado, no período compreendido entre outubro de 2015 e março de 2016. A construção das representações mentais dos atores sociais se dá pelas interações e representações sobre a realidade, aliadas às suas relações coletivas e suas experiências ao longo do tempo. Pode-se concluir que está em curso um processo de padronização racial dos bovinos para atender à demanda do mercado da carne, por outro lado a ovinocultura permanece como atividade secundária, com pouco ou nenhum interesse na comercialização dos animais. De modo geral, convenções de confiança, fidelidade e colaboração apresentam-se como relevantes para a determinação dos canais de comercialização, de modo que a melhor venda é aquela realizada na propriedade rural com o pagamento no ato da venda. Em essência, as convenções que orientam as formas de comercialização são constituídas de diferentes maneiras, sobretudo como fruto da valoração diferenciada que os atores lançam sobre os processos de constituição dos mercados. Os principais canais de comercialização para o terneiro/novilho são o corretor 43,6% e o terminador 41%, o reprodutor comumente é adquirido por meio da troca com outro pecuarista familiar. Os tipos de mercados que sustentaram a análise merecem ajustes à realidade da pecuária familiar, em que é possível haver cruzamentos entre os tipos identificados, visto que não se constituem de relações lineares. A análise mostrou que o pecuarista familiar faz acessa mais de uma canal de comercialização para uma mesma categoria animal, em que os elementos de regulação são reproduzidos nesses diferentes mercados. / The markets of family livestock can be understood as conventions, since they explain the coordination of the actions of the different actors, since they contemplate the diversity of dynamics that can exist in the different markets. It can be said that, if not actively planned and constituted by the actors themselves, market relations are at least mediated by them, for while some distract their production processes and market work, others engage in them. For this reason, it is not enough to understand why certain actors make decisions, it is necessary to understand why they do so and the consequences of those decisions. The objective of this thesis is to understand how the markets of family farming are constituted by analyzing the characteristics that guide the interactions between the social actors of the different marketing channels and the identification and characterization of the different types of markets and the reasons for structure. Aiming at the study of family livestock, interviews were conducted with 39 family ranchers in the municipalities of Bagé, Dom Pedrito and Pinheiro Machado, between October 2015 and March 2016. The construction of the mental representations of social actors is done by the interactions and representations about reality, allied to their collective relations and their experiences over time. It can be concluded that a process of racial standardization of cattle is in progress to meet the demand of the meat market. On the other hand, sheep farming remains a secondary activity with little or no interest in the commercialization of the animals. In general terms, trust, loyalty and collaboration agreements are relevant for the determination of the marketing channels, so that the best sale is the one held at the rural property with the payment at the time of sale. In essence, the conventions that guide the forms of commercialization are constituted in different ways, mainly as a result of the differentiated valuation that the actors throw on the processes of constitution of the markets. The main marketing channels for the calf / steer are the 43.6% broker and the 41% terminator, the breeder is commonly purchased through the exchange with another family farmer. The types of markets that supported the analysis deserve adjustments to the reality of family livestock, where there may be crosses between identified types, since they are not linear relationships. The analysis showed that the family farmer makes access to more than one marketing channel for the same animal category, in which the regulatory elements are reproduced in these different markets.

Bakgrund: Lantbruket står inför stora framtida utmaningar som t.ex. stora befolkningsökningar och minskade jordbruksarealer. Ett förslag för att lösa en del av problemet och öka effektiviteten inom lantbruket är att implementera och använda olika digitala tjänster och produkter. Ett av koncepten för den digitala tekniken som har lyfts fram av bland annat EU är Smart farming. Konceptet är brett och innefattar många olika tekniska lösningar, varav en av dessa är kameraövervakning av boskapsdjur. Frågeställning: Har de lantbrukare som använder kameraövervakning av boskapsdjur upplevt en förändring av sin livskvalitet och har användningen av tekniken medfört några ekonomiska effekter? Metod: Ett kvalitativt angreppssätt valdes där sex lantbrukare intervjuades i semistrukturerade intervjuer. Resultat: De medverkande lantbrukarna var överlag positiva till kameraövervakningssystemen och flera av lantbrukarna anser att tekniken möjliggjort att de kunnat spara tid och att djur i viss utsträckning kunnat räddats. Flera av lantbrukarna menar också att användandet av kameraövervakningssystemen lett till positiva effekter gällande deras livskvalitet, där det framför allt var möjligheten att kunna spendera mer tid med familj och ökad flexibilitet som var bidragande. Slutsatser: Undersökningens resultat tyder på att kameraövervakningssystemen kan spara tid och pengar för lantbrukarna och att den upplevda livskvaliteten förbättras. Den grupp respondenter i undersökningen som upplevde störst effekter av kameraövervakningen var mindre lantbruk. Kameraövervakningen användes till flera olika ändamål, t.ex. brottsförebyggande, kalvning, personalsäkerhet, övervakning av foderbord, lösdrift samt gård och ägor.

The observation, control and the maintenance of the physical condition of sows in acceptable levels are critical to maintain the animal welfare and production in appropriate standards. Lameness causes pain making locomotion difficult. However, lameness is a common disorder in sows that causes negative impacts in both welfare and production. Since the animals that demonstrate this problem, have a smaller number of born-alive piglets, fewer gestation per year and are removed from the herd at a younger age than the ideal. In addition, it is industry practice to limit feed sows to ensure that they remain at an ideal condition score. It is known that, during gestation, each sow should receive a different amount of food according to its body condition. Underweight animals have nutritional deficiency and lower number of piglets per litter. On the other hand, overweight sows have an abnormal development of mammary glands, reducing the amount of milk produced during lactation, causing economic losses. However, moving sows to group gestation makes it difficult to monitor condition score in gestating sows. Both the detection of lameness and the classification of body condition are currently assessed using subjective methods, which is time consuming and difficult to accurately complete. Therefore, the early recognition of animals that present physical condition outside the standards is important to prevent production losses caused by both the aggravation of the conditions presented and the impact on the animals\' welfare. The objective of this project is to obtain three characteristics (body condition score, mass and backfat thickness) through depth images, that proved to be effective on the acquisition of these features in other animals (boars and cows). The second objective is to develop a method for early detection of lameness using the kinematic approach, that has been generating good results and which difficulties have the potential to be reduced by using depth images instead of the method of reflective markers currently used. To predict body condition, a multiple linear regression was obtained using the minor axis of the ellipse fitted around sow\'s body, the width at shoulders, and the angle, of the last rib\'s curvature. To predict backfat, a multiple linear regression was performed using the height of last rib\'s curvature, the perimeter of sow\'s body, the major axis of the ellipse fitted around sow\'s body, the length from snout to rump, and the predicted body condition score. It was possible to obtain the body mass with a simple linear regression using the projected volume of the sows\' body. For lameness detection, three models presented the best accuracy (76.9%): linear discriminant analysis, fine 1-nearest neighbor, and weighted 10-nearest neighbors. The input variables used on the models were obtained from depth videos (number, time, and length of steps for each of the four regions analyzed - left and right shoulders and left and right hips; total walk time; and number of local maxima for head region). As a result of these studies, it has been demonstrated that a depth camera can be used to automate the weight, condition score, backfat thickness, and lameness acquisition/detection in gestating and lactating sows. / A observação, o controle e a manutenção das condições físicas de matrizes suínas em níveis aceitáveis são fundamentais para manter o bem-estar animal e a produção em padrões adequados. A claudicação causa dor e dificuldade de locomoção e, no entanto, é uma desordem comum em matrizes suínas que, além do impacto negativo no bem-estar, gera, também, grandes impactos na produção, uma vez que os animais que demonstram esse problema, apresentam um menor número de leitões nascidos vivos, menor número de partos por ano e são removidas do rebanho a uma idade mais jovem do que a ideal. Sabe-se, ainda, que, durante a gestação, cada matriz deve receber uma quantidade de ração diferenciada de acordo com sua condição corporal. Animais abaixo do peso apresentam deficiência nutricional e menor número de leitões nascidos por ninhada. Já as matrizes com excesso de peso apresentam um desenvolvimento anormal das glândulas mamárias, reduzindo a quantidade de leite produzida durante a lactação, acarretando em perdas econômicas. Tanto a detecção da claudicação quanto a classificação da condição corporal são feitos por meios subjetivos e dependentes da opinião pessoal do tratador, o que pode gerar divergências entre as classificações dadas por cada indivíduo. Destaca-se, portanto, a importância do reconhecimento precoce de animais que apresentam condições físicas fora dos padrões exigidos, visando a prevenção de perdas produtivas causadas tanto pelo agravamento das condições apresentadas quanto pelo grande impacto no bem-estar dos animais. Tendo-se isso em vista, o presente trabalho visou obter três características (escore de condição corporal, massa corporal e espessura de toucinho) por meio de imagens em profundidade, que se mostraram eficazes na obtenção dessas características em outros animais (suínos machos não- castrados e vacas leiteiras). Além disso, buscou-se desenvolver um método para a detecção precoce de claudicação em matrizes suínas, utilizando-se a abordagem da cinemática dos animais, que vem dando bons resultados e cujas dificuldades têm potencial para serem sanadas por meio do uso de imagens em profundidade em vez do método de marcadores reflexivos utilizado atualmente. Para predizer a condição corporal, uma regressão linear múltipla foi obtida usando o menor eixo da elipse ajustada ao redor do corpo da matriz suína, a largura dos ombros e o ângulo da curvatura da última costela. Para predizer a espessura de toucinho, foi realizada uma regressão linear múltipla usando a altura curvatura da última da costela, o perímetro do corpo da matriz, o maior eixo da elipse ajustada, o comprimento do focinho à cauda e o escore predito da condição corporal. Foi possível obter a massa corporal com uma regressão linear simples usando o volume projetado do corpo das matrizes. Para detecção de claudicação, três modelos apresentaram a melhor precisão (76,9%): análise discriminante linear, 1 vizinho mais próximo e 10 vizinhos mais próximos. As variáveis de entrada utilizadas nos modelos foram obtidas a partir de vídeos em profundidade (número, tempo e comprimento de passos para cada uma das quatro regiões analisadas-ombros esquerdo e direito e quadris esquerdo e direito; tempo total de caminhada e número de máximos locais para a região da cabeça). Como resultado desses estudos, observou-se que câmeras em profundidade podem ser utilizadas na automação de medidas de peso, condição corporal, espessura de toucinho e claudicação de matrizes suínas.