Automatic Lameness Detection in a Milking Robot : Instrumentation, measurement software, algorithms for data analysis and a neural network model

The aim of this thesis is to develop a fully automatic lameness detection system that operates in a milking robot. The instrumentation, measurement software, algorithms for data analysis and a neural network model for lameness detection were developed. Automatic milking has become a common practice in dairy husbandry, and in the year 2006 about 4000 farms worldwide used over 6000 milking robots. There is a worldwide movement with the objective of fully automating every process from feeding to milking. Increase in automation is a consequence of increasing farm sizes, the demand for more efficient production and the growth of labour costs. As the level of automation increases, the time that the cattle keeper uses for monitoring animals often decreases. This has created a need for systems for automatically monitoring the health of farm animals. The popularity of milking robots also offers a new and unique possibility to monitor animals in a single confined space up to four times daily. Lameness is a crucial welfare issue in the modern dairy industry. Limb disorders cause serious welfare, health and economic problems especially in loose housing of cattle. Lameness causes losses in milk production and leads to early culling of animals. These costs could be reduced with early identification and treatment. At present, only a few methods for automatically detecting lameness have been developed, and the most common methods used for lameness detection and assessment are various visual locomotion scoring systems. The problem with locomotion scoring is that it needs experience to be conducted properly, it is labour intensive as an on-farm method and the results are subjective. A four balance system for measuring the leg load distribution of dairy cows during milking in order to detect lameness was developed and set up in the University of Helsinki Research farm Suitia. The leg weights of 73 cows were successfully recorded during almost 10,000 robotic milkings over a period of 5 months. The cows were locomotion scored weekly, and the lame cows were inspected clinically for hoof lesions. Unsuccessful measurements, caused by cows standing outside the balances, were removed from the data with a special algorithm, and the mean leg loads and the number of kicks during milking was calculated. In order to develop an expert system to automatically detect lameness cases, a model was needed. A probabilistic neural network (PNN) classifier model was chosen for the task. The data was divided in two parts and 5,074 measurements from 37 cows were used to train the model. The operation of the model was evaluated for its ability to detect lameness in the validating dataset, which had 4,868 measurements from 36 cows. The model was able to classify 96% of the measurements correctly as sound or lame cows, and 100% of the lameness cases in the validation data were identified. The number of measurements causing false alarms was 1.1%. The developed model has the potential to be used for on-farm decision support and can be used in a real-time lameness monitoring system.

Egg production in furnished cages

In the European Union, conventional cages for laying hens will be faded out at the beginning of 2012. The rationale behind this is a public concern over animal welfare in egg production. As alternatives to conventional cages, the European Union Council Directive 1999/74/EC allows non-cage systems and enriched (furnished) cages. Layer performance, behavior, and welfare in differently sized furnished cages have been investigated quite widely during recent decades, but nutrition of hens in this production system has received less attention. This thesis aims to compare production and feed intake of laying hens in furnished and conventional cages and to study the effects of different dietary treatments in these production systems, thus contributing to the general knowledge of furnished cages as an egg production system. A furnished cage model for 8 hens was compared with a 3-hen conventional cage. Three consecutive experiments each studied one aspect of layer diet: The first experiment investigated the effects of dietary protein/energy ratio, the second dietary energy levels, and the third the effects of extra limestone supplementation. In addition, a fourth experiment evaluated the effects of perches on feed consumption and behavior of hens in furnished cages. The dietary treatments in experiments 1 3 generally had similar effects in the two cage types. Thus, there was no evidence supporting a change in nutrient requirements for laying hens when conventional cages are replaced with small-group furnished cages. Moreover, the results from nutritional experiments conducted in conventional cages can be applied to small-group furnished cage systems. These results support the view that production performance comparable with conventional cages can be achieved in furnished cages. All of the advantages of cages for bird welfare are sustained in the small-group furnished cages used here. In addition, frequent use of perches and nests implies a wider behavioral repertoire in furnished cages than in conventional cages. The increase observed in bone ash content may improve bird welfare in furnished cages. The presence of perches diminished feed consumption during the prelaying period and enhanced the feed conversion ratio during the early laying period in furnished cages. However, as the presence or absence of perches in furnished cages had no significant effect on feed consumption after the prelaying period, the lower feed consumption observed in furnished cages than in conventional cages could be attributed to other factors, such as the presence of wood shavings or a nest box. The wider feed trough space per hen in conventional than in furnished cages may partly explain the higher feed consumption observed in conventional cages.

Characterization of north-eastern Australian environments using APSIM for increasing rainfed maize production

Recurring water stresses are a major risk factor for rainfed maize cropping across the highly diverse agro-ecological environments of Queensland (Qld) and northern New South Wales (NNSW). Enhanced understanding of such agro-ecological diversity is necessary to more consistently sample target production environments for testing and targeting release of improved germplasm, and to improve the efficiency of the maize pre-breeding and breeding programs of Qld and New South Wales. Here, we used the Agricultural Production Systems Simulator (APSIM) – a well validated maize crop model to characterize the key distinctive water stress patterns and risk to production across the main maize growing regions of Qld and NNSW located between 15.8° and 31.5°S, and 144.5° and 151.8°E. APSIM was configured to simulate daily water supply demand ratios (SDRs) around anthesis as an indicator of the degree of water stress, and the final grain yield. Simulations were performed using daily climatic records during the period between 1890 and 2010 for 32 sites-soils in the target production regions. The runs were made assuming adequate nitrogen supply for mid-season maize hybrid Pioneer 3153. Hierarchical complete linkage analyses of the simulated yield resulted in five major clusters showing distinct probability distribution of the expected yields and geographic patterns. The drought stress patterns and their frequencies using SDRs were quantified using multivariate statistical methods. The identified stress patterns included no stress, mid-season (flowering) stress, and three terminal stresses differing in terms of severity. The combined frequency of flowering and terminal stresses was highest (82.9%), mainly in sites-soils combinations in the west of Qld and NNSW. Yield variability across the different sites-soils was significantly related to the variability in frequencies of water stresses. Frequencies of water stresses within each yield cluster tended to be similar, but different across clusters. Sites-soils falling within each yield cluster therefore could be treated as distinct maize production environments for testing and targeting newly developed maize cultivars and hybrids for adaptation to water stress patterns most common to those environments.

The critical role of extreme heat for maize production in the United States

Statistical studies of rainfed maize yields in the United States(1) and elsewhere(2) have indicated two clear features: a strong negative yield response to accumulation of temperatures above 30 degrees C (or extreme degree days (EDD)), and a relatively weak response to seasonal rainfall. Here we show that the process-based Agricultural Production Systems Simulator (APSIM) is able to reproduce both of these relationships in the Midwestern United States and provide insight into underlying mechanisms. The predominant effects of EDD in APSIM are associated with increased vapour pressure deficit, which contributes to water stress in two ways: by increasing demand for soil water to sustain a given rate of carbon assimilation, and by reducing future supply of soil water by raising transpiration rates. APSIM computes daily water stress as the ratio of water supply to demand, and during the critical month of July this ratio is three times more responsive to 2 degrees C warming than to a 20% precipitation reduction. The results suggest a relatively minor role for direct heat stress on reproductive organs at present temperatures in this region. Effects of elevated CO2 on transpiration efficiency should reduce yield sensitivity to EDD in the coming decades, but at most by 25%.

Northern challanges for profitable production of quality naked oat

Naked oat (Avena sativa f.sp. nuda L.) is the highest quality cereal in northern growing conditions. However the cultivation area of naked oat is remarkably small. Major challenges for naked oat production are to observe its nakedness. The caryopsis of naked oat is sensitive to mechanical damage at harvest, especially at high grain moisture content. The greater the grain moisture content of naked oat at harvest, the more loses of germination capacity was caused by threshing. For producing high quality naked oat seed, it is recommended that harvesting be done at as low grain moisture content as possible. However, if this is not possible, better germination can be ensure with gentle harvest by reducing the cylinder speed. In spite of conventional oat s excellent fat and amino acid composition in animal feed use, as far as nutritional value is concerned, the total energy yield of oat is weaker than other cereals because of the hulls. Also with naked oat the dehulling is not complete, while hull content on different cultivars mostly varied between one to six percent. In addition to genotype, environmental conditions markedly control the expression of nakedness. Thresher settings had only limited effects on hull content. The function of hulls is to protect the groat, but this was confirmed only for Finnish, small grain, cultivar Lisbeth. The oat kernel is generally covered with fine silky hairs termed trichomes. The trichomes of naked oat are partly lost during threshing and handling of grains. Trichomes can cause itchiness in those handling the grains and also accumulate and form fine dust and can block-up machinery. The cultivars differed considerably in pubescence. Some thresher settings, including increased cylinder speed, slightly increased grain polishing such that grains had some areas completely free of trichomes. Adjusting thresher settings was generally not an efficient means of solving the problems associated with naked oat trichomes. The main differences in cultivation costs between naked and conventional oat lie in the amount of seeds required and the drying costs. The main differences affecting the economic result lie in market prices, yield level and feed value. The results indicate that naked oat is financially more profitable than conventional oat, when the crop is sold at a specific price at all yield levels and when the crop is used as feed at highest yield level. At lower yield levels, conventional oat is, in spite of its lower feed value, the more profitable option for feed use. Dehulled oat did not achieve the same economic result as naked oat, as the cost of dehulling, including the hull waste, was considerable. According to this study naked oat can be cultivated successfully under northern conditions, when taking into consideration the soft, naked grain through cultivation chain.

Influence of harvesting strategy on nutrient supply and production of dairy cows consuming diets based on grass and red clover silage

The main objective of this thesis was to elucidate the effects of regrowth grass silage and red clover silage on nutrient supply and milk production of dairy cows as compared with primary growth grass silages. In the first experiment (publication I), two primary growth and four regrowth grass silages were harvested at two stages of growth. These six silages were fed to 24 lactating dairy cows with two levels of concentrate allowance. Silage intake and energy corrected milk yield (ECM) responses, and the range in these response variables between the diets, were smaller when regrowth silages rather than primary growth silages were fed. Milk production of dairy cows reflected the intake of metabolizable energy (ME), and no differences in the ME utilization were found between the diets based on silages harvested from primary growth and regrowth. The ECM response to increased concentrate allowance was, on average, greater when regrowth rather than primary growth silages were fed. In the second experiment (publication II), two silages from primary growth and two from regrowth used in I were fed to rumen cannulated lactating dairy cows. Cows consumed less feed dry matter (DM), energy and protein, and produced less milk, when fed diets based on regrowth silages rather than primary growth silages. Lower milk production responses of regrowth grass silage diets were mainly due to the lower silage DM intake, and could not be accounted for by differences in energy or protein utilization. Regrowth grass silage intake was not limited due to neutral detergent fibre (NDF) digestion or rumen fill or passage kinetics. However, lower intake may be at least partly attributable to plant diseases such as leaf spot infections, dead deteriorating material or abundance of weeds, which are all higher in regrowth compared with primary growth, and increase with advancing regrowth. In the third experiment (publications III and IV), red clover silages and grass silages harvested at two stages of growth, and a mixed diet of red clover and grass silages, were fed to five rumen cannulated lactating dairy cows. In spite of the lower average ME intake for red clover diets, the ECM production remained unchanged suggesting more efficient utilisation of ME for red clover diets compared with grass diets. Intake of N, and omasal canal flows of total non-ammonia N (NAN), microbial and non-microbial NAN were higher for red clover than for grass silage diets, but were not affected by forage maturity. Delaying the harvest tended to decrease DM intake of grass silage and increase that of red clover silage. The digestion rate of potentially digestible NDF was faster for red clover diets than for grass silage diets. Delaying the harvest decreased the digestion rate for grass but increased it for red clover silage diets. The low intake of early-cut red clover silage could not be explained by silage digestibility, fermentation quality, or rumen fill but was most likely related to the nutritionally suboptimal diet composition because inclusion of moderate quality grass silage in mixed diet increased silage DM intake. Despite the higher total amino acid supply of cows fed red clover versus grass silage diets, further milk production responses on red clover diets were possibly compromised by an inadequate supply of methionine as evidenced by lower methionine concentration in the amino acid profile of omasal digesta and plasma. Increasing the maturity of ensiled red clover does not seem to affect silage DM intake as consistently as that of grasses. The efficiency of N utilization for milk protein synthesis was lower for red clover diets than for grass diets. It was negatively related to diet crude protein concentration similarly to grass silage diets.

Rearing strategies of young dairy calves in relation to production, behaviour and welfare

Vasikoiden kasvatus yksilökarsinoissa, imemismahdollisuuden puute maitojuoton yhteydessä sekä pienet juomamäärät ovat tekijöitä, jotka mahdollisesti voivat vähentää vasikoiden hyvinvointia. Vasikoiden kasvatukseen etsitäänkin uusia tapoja, joissa eläinten käyttäytymistarpeet ja hyvinvointi otetaan entistä paremmin huomioon. Tässä väitöskirjassa vasikoiden kasvatusta on tarkasteltu sekä tuotannon että vasikoiden käyttäytymisen ja hyvinvoinnin kannalta. Väitöskirja koostuu kolmesta kokeesta, joista ensimmäisessä tutkittiin vasikoiden kasvatusta ryhmäkarsinoissa ulkona tai sisällä ja vasikoiden kasvatusta sisällä ryhmä- tai yksilökarsinoissa. Toisessa kokeessa vasikoiden annettiin imeä emiään rajoitetusti lypsyn jälkeen viiden tai kahdeksan viikon ajan ja kolmannessa selvitettiin vasikoiden veden juontia, kun vasikat saivat juomarehua vapaasti. Lisäksi kokeiden yhdistetystä aineistosta analysoitiin eri rehujen syöntimäärien suhdetta sekä rehujen vaikutusta kasvuun ennen ja jälkeen maidosta vieroituksen. Tutkimuksessa todettiin, että vasikoita voi ryhmässä kasvattaa kylmissä ja vaihtelevissa sääoloissa ulkona, kunhan ne hoidetaan ja ruokitaan erittäin huolellisesti. Kylmällä ilmalla vasikat saattavat kuitenkin syödä väkirehua vähemmän varsinkin, jos ruokailupaikka on ulkona ja makuualue sisällä. Ryhmässä kasvaneet vasikat aloittivat sekä kuivien rehujen syönnin että märehtimisen nuorempina kuin yksilökarsinassa kasvaneet. Ryhmissä esiintyvää käyttäytymisongelmaa, toisten vasikoiden imemistä, voidaan vähentää hoito- ja ruokintamenetelmillä. Annettaessa vasikoiden imeä emiään rajoitetusti lypsyn jälkeen vasikat oppivat imemään emiään hyvin nopeasti. Lypsytyö vaikeutui muutamien lehmien kohdalla, sillä ne pidättivät maitoa lypsettäessä. Saadessaan imeä rajoitetusti vasikat imivät suurehkoja maitomääriä kerrallaan. Vieroittaminen suurilta maitomääriltä viiden viikon iässä oli kuitenkin liian aikaista, koska vasikat eivät vielä syöneet riittävästi kuivia rehuja. Vieroitus emästä niin viiden kuin kahdeksankin viikon iässä aiheutti vasikoissa levottomuutta ja ääntelyn lisääntymistä. Saadessaan hapatettua juomarehua vapaasti vasikat joivat keskimäärin vain vähän vettä, olipa vesilähteenä avoin ämpäri tai vesinippa. Vasikoiden välillä oli suurta vaihtelua veden juontimäärissä. Viikkoa ennen maidosta vieroitusta vasikat joivat 0-3 l vettä päivässä. Vasikat joivat nipasta kerrallaan vähemmän vettä kuin ämpäristä, ja käyttivät enemmän aikaa päivässä veden juomiseen kuin vesiämpäristä juoneet vasikat. Suurin osa vasikoista joi vettä juomanipoista erikoisella tavalla esimerkiksi painamalla nippaa otsalla ja juomalla tippuvaa vettä. Vesinipat voivat olla siis vasikoille joko vaikeita tai epämukavia käyttää. Vasikoiden juoman maitomäärän lisääntyessä kasvu lisääntyy selvästi. Runsas maidon juominen vähentää kuitenkin kuivien rehujen syöntiä ja vieroitusvaiheessa kasvu voi hidastua. Vasikat olisikin tärkeää vieroittaa vähitellen, ettei muutoksia kasvuun tulisi. Syönti- ja kasvutulokset eivät aina anna oikeaa kuvaa kasvatusmenetelmien eroista eläinten hyvinvoinnin kannalta. Käyttäytyminen on herkkä hyvinvoinnin mittari ja se tulisikin aina huomioida eri kasvatusmenetelmiä arvioitaessa. .

Understanding the organic chain : The framework of the interaction between the actors in organic chains in relation to ecological modernisation of food production

The sustainability of food production has increasingly attracted the attention of consumers, farmers, food and retailing companies, and politicians. One manifestation of such attention is the growing interest in organic foods. Organic agriculture has the potential to enhance the ecological modernisation of food production by implementing the organic method as a preventative innovation that simultaneously produces environmental and economic benefits. However, in addition to the challenges to organic farming, the small market share of organic products in many countries today and Finland in particular risks undermining the achievement of such benefits. The problems identified as hindrances to the increased consumption of organic food are the poor availability, limited variety and high prices of organic products, the complicated buying decisions and the difficulties in delivering the intangible value of organic foods. Small volumes and sporadic markets, high costs, lack of market information, as well as poor supply reliability are obstacles to increasing the volume of organic production and processing. These problems shift the focus from a single actor to the entire supply chain and require solutions that involve more interaction among the actors within the organic chain. As an entity, the organic food chain has received very little scholarly attention. Researchers have mainly approached the organic chain from the perspective of a single actor, or they have described its structure rather than the interaction between the actors. Consequently, interaction among the primary actors in organic chains, i.e. farmers, manufacturers, retailers and consumers, has largely gone unexamined. The purpose of this study is to shed light on the interaction of the primary actors within a whole organic chain in relation to the ecological modernisation of food production. This information is organised into a conceptual framework to help illuminate this complex field. This thesis integrates the theories and concepts of three approaches: food system studies, supply chain management and ecological modernisation. Through a case study, a conceptual system framework will be developed and applied to a real life-situation. The thesis is supported by research published in four articles. All examine the same organic chains through case studies, but each approaches the problem from a different, complementary perspective. The findings indicated that regardless of the coherent values emphasising responsibility, the organic chains were loosely integrated to operate as a system. The focus was on product flow, leaving other aspects of value creation largely aside. Communication with consumers was rare, and none of the actors had taken a leading role in enhancing the market for organic products. Such a situation presents unsuitable conditions for ecological modernisation of food production through organic food and calls for contributions from stakeholders other than those directly involved in the product chain. The findings inspired a revision of the original conceptual framework. The revised framework, the three-layer framework , distinguishes the different layers of interaction. By gradually enlarging the chain orientation the different but interrelated layers become visible. A framework is thus provided for further research and for understanding practical implications of the performance of organic food chains. The revised framework provides both an ideal model for organic chains in relation to ecological modernisation and demonstrates a situation consistent with the empirical evidence.

Microclimate and gas emissions in dairy buildings : Instrumentation, theory and measurements

The aim of this thesis was to develop measurement techniques and systems for measuring air quality and to provide information about air quality conditions and the amount of gaseous emissions from semi-insulated and uninsulated dairy buildings in Finland and Estonia. Specialization and intensification in livestock farming, such as in dairy production, is usually accompanied by an increase in concentrated environmental emissions. In addition to high moisture, the presence of dust and corrosive gases, and widely varying gas concentrations in dairy buildings, Finland and Estonia experience winter temperatures reaching below -40 ºC and summer temperatures above +30 ºC. The adaptation of new technologies for long-term air quality monitoring and measurement remains relatively uncommon in dairy buildings because the construction and maintenance of accurate monitoring systems for long-term use are too expensive for the average dairy farmer to afford. Though the documentation of accurate air quality measurement systems intended mainly for research purposes have been made in the past, standardised methods and the documentation of affordable systems and simple methods for performing air quality and emissions measurements in dairy buildings are unavailable. In this study, we built three measurement systems: 1) a Stationary system with integrated affordable sensors for on-site measurements, 2) a Wireless system with affordable sensors for off-site measurements, and 3) a Mobile system consisting of expensive and accurate sensors for measuring air quality. In addition to assessing existing methods, we developed simplified methods for measuring ventilation and emission rates in dairy buildings. The three measurement systems were successfully used to measure air quality in uninsulated, semi-insulated, and fully-insulated dairy buildings between the years 2005 and 2007. When carefully calibrated, the affordable sensors in the systems gave reasonably accurate readings. The spatial air quality survey showed high variation in microclimate conditions in the dairy buildings measured. The average indoor air concentration for carbon dioxide was 950 ppm, for ammonia 5 ppm, for methane 48 ppm, for relative humidity 70%, and for inside air velocity 0.2 m/s. The average winter and summer indoor temperatures during the measurement period were -7º C and +24 ºC for the uninsulated, +3 ºC and +20 ºC for the semi-insulated and +10 ºC and +25 ºC for the fully-insulated dairy buildings. The measurement results showed that the uninsulated dairy buildings had lower indoor gas concentrations and emissions compared to fully insulated buildings. Although occasionally exceeded, the ventilation rates and average indoor air quality in the dairy buildings were largely within recommended limits. We assessed the traditional heat balance, moisture balance, carbon dioxide balance and direct airflow methods for estimating ventilation rates. The direct velocity measurement for the estimation of ventilation rate proved to be impractical for naturally ventilated buildings. Two methods were developed for estimating ventilation rates. The first method is applicable in buildings in which the ventilation can be stopped or completely closed. The second method is useful in naturally ventilated buildings with large openings and high ventilation rates where spatial gas concentrations are heterogeneously distributed. The two traditional methods (carbon dioxide and methane balances), and two newly developed methods (theoretical modelling using Fick s law and boundary layer theory, and the recirculation flux-chamber technique) were used to estimate ammonia emissions from the dairy buildings. Using the traditional carbon dioxide balance method, ammonia emissions per cow from the dairy buildings ranged from 7 g day-1 to 35 g day-1, and methane emissions per cow ranged from 96 g day-1 to 348 g day-1. The developed methods proved to be as equally accurate as the traditional methods. Variation between the mean emissions estimated with the traditional and the developed methods was less than 20%. The developed modelling procedure provided sound framework for examining the impact of production systems on ammonia emissions in dairy buildings.