O presente trabalho teve como objetivo apresentar uma metodologia que permita encontrar o posicionamento e rastrear as diferentes localizações de um suíno em uma baia, utilizando o valor do Receiver Signal Strenght Indicator (RSSI), entre o dispositivo móvel (suíno) e três dispositivos fixos, e uma Rede Neural Artificial do tipo Perceptron Multicamadas (PMC), responsável por interpretar os sinais RSSI e transformá-los em valores conhecidos, como em um plano cartesiano, com coordenadas no eixo X e eixo Y. A região de teste foi dividida em 289 pontos, sendo 286 utilizados para coleta de dados e para o treinamento da rede PMC. Para cada ponto, foram armazenados a sua posição dentro da baia e o valor RSSI entre o dispositivo móvel e os três dispositivos fixos. O processo foi repetido para 8 pontos escolhidos aleatoriamente dentro do espaço de teste e inseridos como entradas na rede PMC. Após treinamentos e operações realizadas com diversas arquiteturas foi possível concluir que àquela dotada de 10 neurônios na camada intermediária consistiu na melhor alternativa, cujos resultados de monitoramento e rastreamento das posições do dispositivo móvel foram encontradas com valores aceitáveis de exatidão. / This paper aims to present a methodology to find the positioning and tracking of the different locations of a pig in a stall, using the value of the Receiver Signal Strength Indicator (RSSI), between the mobile device (pig) and three devices fixed, and an Artificial Neural Network Multilayer Perceptron (MLP), responsible for interpreting the RSSI signals and turning them into known values, such as on a Cartesian plane, with coordinates on X axis and Y axis. The test region was divided into 289 points, with 286 points used for data collection and training of PMC network, and for each point, it was stored its position inside the stall and its RSSI value between the mobile devices and the three fixed. The process was repeated for 8 points chosen randomly within the space of test and entered as inputs into the PMC network. After training and operations with various architectures it was concluded that the architecture with 10 neurons in the hidden layer was the best alternative, whose the results of monitoring and tracking the positions of mobile device were found with acceptable accuracy.

O presente trabalho teve como objetivo apresentar uma metodologia que permita encontrar o posicionamento e rastrear as diferentes localizações de um suíno em uma baia, utilizando o valor do Receiver Signal Strenght Indicator (RSSI), entre o dispositivo móvel (suíno) e três dispositivos fixos, e uma Rede Neural Artificial do tipo Perceptron Multicamadas (PMC), responsável por interpretar os sinais RSSI e transformá-los em valores conhecidos, como em um plano cartesiano, com coordenadas no eixo X e eixo Y. A região de teste foi dividida em 289 pontos, sendo 286 utilizados para coleta de dados e para o treinamento da rede PMC. Para cada ponto, foram armazenados a sua posição dentro da baia e o valor RSSI entre o dispositivo móvel e os três dispositivos fixos. O processo foi repetido para 8 pontos escolhidos aleatoriamente dentro do espaço de teste e inseridos como entradas na rede PMC. Após treinamentos e operações realizadas com diversas arquiteturas foi possível concluir que àquela dotada de 10 neurônios na camada intermediária consistiu na melhor alternativa, cujos resultados de monitoramento e rastreamento das posições do dispositivo móvel foram encontradas com valores aceitáveis de exatidão. / This paper aims to present a methodology to find the positioning and tracking of the different locations of a pig in a stall, using the value of the Receiver Signal Strength Indicator (RSSI), between the mobile device (pig) and three devices fixed, and an Artificial Neural Network Multilayer Perceptron (MLP), responsible for interpreting the RSSI signals and turning them into known values, such as on a Cartesian plane, with coordinates on X axis and Y axis. The test region was divided into 289 points, with 286 points used for data collection and training of PMC network, and for each point, it was stored its position inside the stall and its RSSI value between the mobile devices and the three fixed. The process was repeated for 8 points chosen randomly within the space of test and entered as inputs into the PMC network. After training and operations with various architectures it was concluded that the architecture with 10 neurons in the hidden layer was the best alternative, whose the results of monitoring and tracking the positions of mobile device were found with acceptable accuracy.

<p dir="ltr">Agricultural applications often require human experts with domain knowledge to ensure compliance and improve productivity, which can be costly and inefficient. To tackle this problem, automated video systems can be implemented for agricultural tasks thanks to the ubiquity of cameras. In this thesis, we focus on designing and implementing video analytics systems for real applications in agriculture by combining both traditional image processing and recent advancements in computer vision. Existing research and available methods have been heavily focused on obtaining the best performance on large-scale benchmarking datasets, while neglecting the applications to real-world problems. Our goal is to bridge the gap between state-of-art methods and real agricultural applications. More specifically, we design video systems for the two tasks of monitoring turkey behavior for turkey welfare and handwashing action recognition for improved food safety. For monitoring turkeys, we implement a turkey detector, a turkey tracker, and a turkey head tracker by combining object detection and multi-object tracking. Furthermore, we detect turkey activities by incorporating motion information. For recognizing handwashing activities, we combine a hand extraction method for focusing on the hand regions with a neural network to build a hand image classifier. In addition, we apply a two-stream network with RGB and hand streams to further improve performance and robustness.</p><p dir="ltr">Besides designing a robust hand classifier, we explore how dataset attributes and distribution shifts can impact system performance. In particular, distribution shifts caused by changes in hand poses and shadow can cause a classifier’s performance to degrade sharply or breakdown beyond a certain point. To better explore the impact of hand poses and shadow and to mitigate the induced breakdown points, we generate synthetic data with desired variations to introduce controlled distribution shift. Experimental results show that the breakdown points are heavily impacted by pose and shadow conditions. In addition, we demonstrate mitigation strategies to significant performance degradation by using selective additional training data and adding synthetic shadow to images. By incorporating domain knowledge and understanding the applications, we can effectively design video analytics systems and apply advanced techniques in agricultural scenarios.</p>

La dispersion des animaux est un processus clé pour le maintien des populations et métapopulations. En particulier, les amphibiens se déplacent beaucoup durant leur cycle de vie fait de migrations saisonnières. Ceci les rend sensibles à la structure du paysage. En effet, les modifications anthropiques comme l’aménagement d’infrastructures linéaires peuvent constituer des barrières écologiques pour les animaux et limiter le flux de gènes dans le paysage. Si les effets des routes et des lignes chemins de fer sur les déplacements des animaux ont été largement étudiés, les effets des infrastructures liées aux activités de sport d’hiver comme les pistes de ski et les remontées mécaniques sont moins bien connus. Ainsi, la perméabilité des milieux de montagne anthropisés est peu connue. Cette thèse vise à quantifier la perméabilité écologique (ou connectivité fonctionnelle), des domaines skiables pour un amphibien de montagne largement répandu : la Grenouille rousse (Rana temporaria Linnaeus 1758). Pour ce faire, nous combinons un suivi individuel de déplacements et une étude de génétique du paysage pour mieux comprendre les facteurs influençant les déplacements de grenouilles et les flux de gènes dans le paysage. Le suivi individuel est réalisé au moyen de balises GPS et d’analyses de sélection de pas (step selection analysis). L’étude de génétique du paysage est réalisée en prélevant des échantillons d’ADN dans trois domaines skiables de Savoie ainsi que dans une zone témoin située dans le Parc national de la Vanoise. Nous avons testé l’influence de plusieurs variables paysagères liées à la topographie (pente, altitude…), les habitats (réseau de zones humides, couverture du sol…) ou les activités humaines (présence de routes, de pistes de ski) sur le flux de gènes. En utilisant une nouvelle méthode d’optimisation de surface de résistance, nous montrons l’importance du réseau de zones humides pour les amphibiens de montagne. D’une part, elles servent d’habitat préférentiel et d’autre part, un réseau dense de zones humides permet d’assurer une bonne connectivité des paysages pour la Grenouille rousse. Nous mettons également en évidence que l’hétérogénéité des tailles efficaces de populations dans un paysage peut fausser les distances génétiques mesurées entre populations et nous recommandons l’utilisation de techniques pour s’affranchir de tels problèmes. Nos travaux sont importants dans une perspective de gestion durable des domaines skiables, l’aménagement de pistes de ski ou autres infrastructures de loisir pouvant être accompagnée de destructions de zones humides. / Animal disperal is a key process for maintaining populations and metapopulations. In particular, amphibians move a lot during their life cycle made of seasonal migrations. This makes them sensitive to the structure of the landscape. Indeed, anthropogenic changes such as linear infrastructure can act as ecological barriers for animals and limit gene flow within the landscape. While the effects of roads and railways on animal movements have been widely studied, the effects of infrastructure related to winter sports activities such as ski slopes and lifts are less well known. Thus, the permeability of anthropized mountain environments is little known. This thesis aims to quantify the ecological permeability (or functional connectivity) of ski areas for a widely distributed mountain amphibian: the European common frog (Rana temporaria Linnaeus 1758). To do this, we combine individual movement tracking and landscape genetics studies to better understand the factors influencing frog movements and gene flow within the landscape. Individual tracking is carried out by means of GPS tags and step selection analysis. The landscape genetics study is carried out by gathering DNA samples from three ski areas in Savoie and a control area located in the Vanoise National Park. We tested the influence of several landscape variables related to topography (slope, altitude...), habitats (wetland network, land cover...) or human activities (presence of roads, ski slopes) on gene flow. By using a new method of resistance surface optimization, we show the importance of the wetland network for mountain amphibians. On the one hand, they serve as preferential habitat and, on the other hand, a dense network of wetlands ensures good landscape connectivity for the Common frog. We also highlight that the heterogeneity in effective population sizes in a landscape can distort the genetic distances measured between populations and we recommend the use of techniques to overcome such problems. Our work is important for the sustainable management of ski areas, the development of ski slopes or other recreational infrastructure that can be accompanied by the destruction of wetlands.

This diplom work entitled "Monitoring system for the detection of motility and position of laboratory animals after anesthesia" focuses on the design and implementation of non-contact detection of the rat or mouse position in the enclosure with a transparent cover. The aim of this semester paper is to find suitable methods of realizing contactless detection of the position of a laboratory rat or mouse. This automatic positioning of the animal will be used as the basis for controlling the irradiator in the next follow-up work, which will "shade" animal movement and aim at the scar on the animal's back. The rat that is located inside our enclosure is either standard or movable after anesthesia. In this work I first deal with searches of automatic monitoring systems for detecting the position of animals in the enclosure. Then, in the practical part, I test three types of cameras for image detection of rat position. Evaluation software for motion analysis will largely be solved in the follow-up diploma thesis.Project made like monitoring and detecting software based on OpenCV.

Die zunehmende Verfügbarkeit von Satellitenfernerkundungs- und Wildtier-Telemetriedaten eröffnet neue Möglichkeiten für eine verbesserte Überwachung von Wildtierhabitaten durch Habitatmodelle, doch fehlt es häufig an geeigneten Ansätzen, um dieses Potenzial voll auszuschöpfen. Das übergeordnete Ziel dieser Arbeit bestand in der Konzipierung und Weiterentwicklung von Ansätzen zur Nutzung des Potenzials großer Satellitenbild- und Telemetriedatensätze in Habitatmodellen. Am Beispiel von drei großen Säugetierarten in Europa (Eurasischer Luchs, Rothirsch und Reh) wurden Ansätze entwickelt, um (1) Habitatmodelle mit dem umfangreichsten global und frei verfügbaren Satellitenbildarchiv der Landsat-Satelliten zu verknüpfen und (2) Wildtier-Telemetriedaten über Wildtierpopulationen hinweg in großflächigen Analysen der Habitateignung und -nutzung zu integrieren. Die Ergebnisse dieser Arbeit belegen das enorme Potenzial von Landsat-basierten Variablen als Prädiktoren in Habitatmodellen, die es ermöglichen von statischen Habitatbeschreibungen zu einem kontinuierlichen Monitoring von Habitatdynamiken über Raum und Zeit überzugehen. Die Ergebnisse meiner Forschung zeigen darüber hinaus, wie wichtig es ist, die Kontextabhängigkeit der Lebensraumnutzung von Wildtieren in Habitatmodellen zu berücksichtigen, insbesondere auch bei der Integration von Telemetriedatensätzen über Wildtierpopulationen hinweg. Die Ergebnisse dieser Dissertation liefern neue ökologische Erkenntnisse, welche zum Management und Schutz großer Säugetiere beitragen können. Darüber hinaus zeigt meine Forschung, dass eine bessere Integration von Satellitenbild- und Telemetriedaten eine neue Generation von Habitatmodellen möglich macht, welche genauere Analysen und ein besseres Verständnis von Lebensraumdynamiken erlaubt und so Bemühungen zum Schutz von Wildtieren unterstützen kann. / The growing availability of satellite remote sensing and animal tracking data opens new opportunities for an improved monitoring of wildlife habitats based on habitat models, yet suitable approaches for making full use of this potential are commonly lacking. The overarching goal of this thesis was to develop and advance approaches for harnessing the potential of big satellite image and animal tracking data in habitat models. Specifically, using three large mammal species in Europe as an example (Eurasian lynx, red deer, and roe deer), I developed approaches for (1) linking habitat models to the largest global and freely available satellite image record, the Landsat image archive, and (2) for integrating animal tracking datasets across wildlife populations in large-area assessments of habitat suitability and use. The results of this thesis demonstrate the enormous potential of Landsat-based variables as predictors in habitat models, allowing to move from static habitat descriptions to a continuous monitoring of habitat dynamics across space and time. In addition, my research underscores the importance of considering context-dependence in species’ habitat use in habitat models, particularly also when integrating tracking datasets across wildlife populations. The findings of this thesis provide novel ecological insights that help to inform the management and conservation of large mammals and more broadly, demonstrate that a better integration of satellite image and animal tracking data will allow for a new generation of habitat models improving our ability to monitor and understand habitat dynamics, thus supporting efforts to restore and protect wildlife across the globe.

Throughout the animal kingdom, individuals often differ consistently from one another in how they cope with their environment. In particular, consistent behavioural variation, known as animal personality, is a substantial driver of a range of important ecological and evolutionary processes. As most animal species are social for at least part of their lives and group living is common, a crucial link between personality and the social context may be expected. In this thesis I systematically investigate this link, using three-spined stickleback fish (Gasterosteus aculeatus) as my model system. I begin by showing that fish vary consistently in their boldness and sociability, with only boldness being positively linked to food intake for fish at rest. This finding provides support for evolutionary theory that links personality variation to life-history strategies, and lays the basis for work related to the social context. I continue by investigating how the social context may modulate personality variation and show that short contact with a social group may have carry-over effects and obfuscate personality expression when individuals are alone. Next, I observed fish in different pairs over time and found that social experience from both the current as well as previous social contexts are integrated in the risk-taking and leadership decisions of individuals but also depends on their boldness type. This result provides support for the importance of social feedback in the expression of personality differences. I go on to demonstrate that, in a pair, bolder fish have lower social attraction, with positive effects on individual’s leadership but negative effects on social coordination. Finally, by detailed tracking of the collective movements and group foraging of free-swimming shoals, I reveal boldness and sociability have complementary driving effects of on social structure, collective behaviour, and group functioning. Furthermore, I show that in turn the group composition determines the performance of individual personality types, providing a potential adaptive explanation for the maintenance of personality variation. Taken together, these studies provide an integrated account of animal personality and the social context and highlight the presence of a feedback loop between them, with personality variation being a key driver of collective behaviour and group functioning but also strongly affected and potentially maintained by it.

This diploma thesis, entitled "Monitoring System for Determination of Motility and Position of Laboratory Animals After Anesthesia", focuses on the design and implementation of contactless detection of the position of a rat or mouse in an enclosure with a transparent cover. The aim of the semester work is to find suitable methods of realization of contactless detection of rat or mouse position and to automatically determine and display average speed or other movement characteristics. The assignment arose from the needs of animal monitoring after curative intervention and also as a necessary utility for future "shading" animal movement (automatic targeting of the scar on the animal's back). The rat, which is located inside our enclosure, is either moving as standard or is dazed after anesthesia. In this work I deal first with search of automatic monitoring systems for detection of animals in the enclosure. Then in the practical part are tested three types of cameras for visual detection of rat position and a script for automatic detection and analysis of rat movement is designed. The system works like a camera eye which in real time is able to find the area of a black box in its field of view and then limit the detection area to the size of this box and then automatically detects the center of gravity and counts. and evaluates the obtained speed with an average calculated with a test of 10 mice - voices on the screen the mouse status in the previous ten seconds. for no stressed animal The rat that is located inside our enclosure is either standard or movable after anesthesia. In this work I first deal with searches of automatic monitoring systems for detecting the position of animals in the enclosure. Then, in the practical part, I test three types of cameras for image detection of rat position. Evaluation software for motion analysis will largely be solved in the follow-up diploma thesis.Project made like monitoring and detecting software based on OpenCV.

No description available.