This paper examines causal mechanisms of the process leading up to the Russian military intervention in Syria that began in September 2015. It aims to concretize the causal processes of three different hypotheses that are based on commonplace assumptions of Russian foreign policy on Syria. It thoroughly explores three different causal paths, mapping events that may have had implications to the apparent change of heart within the Russian leadership. The paper analyses the relevance of these processes through a rational choice theory framework.

No presente trabalho, foram investigados os efeitos de tratamentos térmicos em vapor d’água, em óxido nítrico e em hidrogênio nas propriedades físico-químicas e elétricas de filmes dielétricos crescidos termicamente e/ou depositadas por sputtering sobre lâminas de carbeto de silício. A caracterização foi realizada antes e após tratamentos térmicos nesses ambientes através de técnicas que utilizam feixes de íons. Em alguns casos, a caracterização elétrica também foi realizada. A investigação da incorporação e distribuição em profundidade de hidrogênio e oxigênio após tratamentos de SiO2/SiC e SiO2/Si em vapor d’água mostrou que há diferenças marcantes na interação da água com as duas estruturas. Observou-se maior incorporação de oxigênio no filme pré-existente de SiO2 sobre o SiC do que em SiO2/Si, evidenciando uma maior concentração de defeitos nos filmes sobre SiC. A incorporação de hidrogênio também foi maior nas estruturas SiO2/SiC, sendo observada em todas as regiões do filme de SiO2 crescido sobre SiC. Nos filmes crescidos sobre Si, no entanto, a incorporação deuse, principalmente, na região da superfície do filme de óxido. A interação do vapor d’água com estruturas SiO2/SiC e SiO2/Si com filmes depositados por sputtering também foi investigada. Foi constatada uma incorporação distinta da observada para essas estruturas quando seus óxidos foram crescidos termicamente. A incorporação de hidrogênio do vapor d’água em estruturas com filmes de SiO2 depositados por sputtering sobre SiC e sobre Si ocorre, principalmente, na interface SiO2/substrato. A distribuição em profundidade de oxigênio após a exposição a vapor d'água a 800°C revelou que ele é incorporado em toda a espessura dos óxidos depositados sobre ambos os substratos, evidenciando a alta mobilidade dos átomos de oxigênio nesses filmes de óxido. O crescimento térmico antes da deposição de SiO2 sobre o SiC levou à incorporação de menores quantidades de hidrogênio, quando comparadas com as estruturas relativas a filmes apenas depositados. No entanto, à medida que o tempo de oxidação térmica foi aumentado, observou-se maior incorporação de hidrogênio, o que foi atribuído à formação de defeitos no filme de óxido susceptíveis à interação com o mesmo. O crescimento térmico por um tempo curto seguido pela deposição de SiO2 e o crescimento térmico não seguido de outro tratamento levaram a menores incorporações de D do que a deposição não seguida de outro tratamento, o que pode ser correlacionado com as melhores características elétricas observadas nessas estruturas. Outro tema abordado foi a incorporação de hidrogênio através de tratamento térmico em 2H2, com e sem a presença de um eletrodo de platina, em filmes dielétricos crescidos em atmosfera de O2, NO e via tratamento térmico sequencial nesses dois gases. Quando o crescimento térmico em O2 foi seguido de tratamento em NO, foi observada uma forte troca isotópica entre o oxigênio da fase gasosa e o oxigênio do filme de SiO2, evidenciando a alta mobilidade dos átomos de oxigênio nesses filmes. A incorporação de hidrogênio mostrou-se fortemente dependente da rota utilizada no crescimento do filme dielétrico. Sem a presença do eletrodo de platina, o crescimento do filme dielétrico direto em NO foi a rota que apresentou a maior incorporação de hidrogênio. A presença de platina, por sua vez, promoveu um aumento na incorporação de hidrogênio nos filmes dielétricos obtidos através das três rotas de crescimento. Em todos os casos, observou-se que a incorporação de hidrogênio ocorre, principalmente, na região da interface entre o filme dielétrico e o SiC. A incorporação de maiores quantidades de hidrogênio foi associada com a presença de N previamente incorporado. A atmosfera reativa utilizada no crescimento térmico dos filmes dielétricos também mostrou influência nas características elétricas das estruturas analisadas. A caracterização por curvas C-V mostrou um aumento no deslocamento da tensão de banda plana após tratamentos térmicos em 2H2, indicando o aumento e/ou formação de carga positiva. Por fim, a interação de vapor d'água em estruturas de SiO2/SiC e de SiO2/Si com filmes crescidos termicamente e tratadas em NO foi investigada. Observou-se que as estruturas SiO2/SiC que foram submetidas a tratamentos térmicos em NO apresentaram menor incorporação de hidrogênio, devido à exposição a vapor d'água. Esse efeito também foi observado em estruturas SiO2/SiC quando o pós-tratamento em NO foi substituído por um póstratamento em argônio na mesma temperatura e tempo, indicando que a temperatura de tratamento é a responsável pelas menores incorporações de hidrogênio, não a reatividade do gás empregado. / In the present work, effects of thermal treatments in water vapor, in nitric oxide, and in hydrogen in the physicochemical and in the electrical properties of dielectric films thermally grown and/or deposited by sputtering on silicon carbide were investigated. The characterization was performed using ion beam analyses before and after thermal treatments in these atmospheres. In some cases, the electrical characterization was also performed. The investigation of the incorporation and depth distribution of hydrogen and oxygen after annealing of SiO2/SiC and SiO2/Si in water vapor evidenced that there are striking differences regarding water interaction with these two structures. It was observed larger oxygen incorporation in the pre-existent SiO2 film on SiC than in the SiO2/Si, which evidences higher concentration of defects in oxide films on SiC. The incorporation of hydrogen was also larger in SiO2/SiC structures, being observed in all regions of the dielectric film. In oxide films grown on Si, however, the incorporation occurred mainly in the surface region of the oxide. The interaction of water vapor with SiO2/SiC and SiO2/Si structures, whose films were deposited by sputtering, was also investigated. A distinct incorporation was observed when comparing results from structures whose oxides were thermally grown. The incorporation of hydrogen from water vapor in structures in which SiO2 films were deposited by sputtering on SiC and on Si, occurred mainly in the SiO2/substrate interface. The oxygen depth distribution after exposure to water vapor at 800°C revealed that it was incorporated in all depths of the oxides that were deposited on both substrates, evidencing the high mobility of oxygen atoms in these oxide films. The thermal growth prior to SiO2 deposition on SiC led to the incorporation of smaller amounts of hydrogen, compared with structures with films that were only deposited. Nevertheless, as the thermal oxidation time increases, a larger incorporation of hydrogen was observed, which was attributed to the formation of defects in the oxide film that are more likely to interact with hydrogen. The thermal growth for short time followed by the deposition of SiO2 and the thermal growth not followed by any other treatment led to lower amounts of hydrogen, when compared with the deposition not followed by another treatment, which can be correlated with the improvement in the electrical characteristics observed in these structures. Another subject investigated was the incorporation of hydrogen by 2H2 anneal, with and without the presence of platinum, in dielectric films thermally grown in O2, NO, and in sequential thermal treatments in these two atmospheres. In the case of thermal growth in O2 followed by NO anneal, it was observed a notable isotopic exchange between oxygen from the gas phase and oxygen from the SiO2 film, evidencing that oxygen atoms are highly mobile in these films. The incorporation of hydrogen was showed to be highly dependent on the route employed in the dielectric film growth, being the direct growth of dielectric films in NO the one that presented larger incorporation without the presence of Pt electrode. The presence of this metal increases the incorporation of hydrogen in all dielectric films. In all cases, it was observed that the incorporation of hydrogen occurred mainly in the interface region between the dielectric film and the SiC. The incorporation of larger amounts of hydrogen was associated with the presence of N that was previously incorporated. The reactive atmosphere employed in the thermal growth of dielectric films also was observed to affect electrical characteristics in analyzed structures. The characterization by C-V measurements evidenced an increase in the flatband voltage shift after annealing in 2H2, indicating the increase and/or the formation of positive charge. Finally, the interaction of water vapor in SiO2/SiC and SiO2/Si structures with dielectric films thermally grown and annealed in NO was investigated. It was observed that SiO2/SiC structures that were submitted to NO anneal presented less hydrogen incorporation due to exposure to water vapor. This behavior was also observed in SiO2/SiC structures when the NO anneal was replaced by an annealing in Ar at the same temperature and time, indicating that the temperature of the annealing was responsible by the less incorporation of hydrogen instead of the reactivity of the gas employed.

INTRODUÇÃO: A poliquimioterapia (PQT-OMS) para tratamento da hanseníase resultou em drástica redução da sua prevalência, mas em limitado impacto na detecção de casos novos (CN), que se manteve estável em Santa Catarina, estado na fase de pós-eliminação. Casos com altos índices baciloscópicos apresentam risco de recidiva tardia e podem consistir em focos persistentes de transmissão da doença. OBJETIVOS: Avaliar a recidiva da doença em amostra de pacientes virchowianos regularmente tratados com PQT-OMS 24 doses; ensaios de resistência terapêutica em camundongos para os casos suspeitos; resposta imune específica (celular e humoral) e detecção do DNA do M. leprae nos casos-índices (CI) e seus contatos indradomiciliares (CID); e os achados frente aos indicadores epidemiológicos e operacionais de Santa Catarina. CASUÍSTICA E MÉTODOS: A partir da busca no Sistema de Informação de Agravos de Notificação (SINAN), foram selecionados 46 CI, tratados entre 1990-2000, e 187 CID, dos municípios de Itajaí e Joinville. A avaliação constituiu de: exames dermatoneurológico e anatomopatológico da pele; baciloscopia do esfregaço cutâneo; reação de Mitsuda; sorologia para glicolipídeo fenólico-I (IgM-anti-PGL-I); ensaios de resistência terapêutica em camundongos e de detecção do DNA bacilar no muco nasal por reação em cadeia da polimerase (PCR) para as seqüências repetitivas específicas RLEP-130 e RLEP-372. RESULTADOS: Entre os CI após alta por cura (m=11,2 ± 3 anos), a idade média (57,3±14,5 anos) foi superior (p<0,05) comparada aos CID, com predomínio de homens (p=0,001) e Mitsuda-negativos (78,6%). Em quatro casos (8,7%) considerados recidivados, o valor sérico médio (0,365) e as freqüências da positividade do anti-PGL-I e a da RLEP-130 (75%; p=0,03) foram consistentemente elevados, e os testes em camundongos, negativos. Dentre os 187 CID, 22 (11,8%) adoeceram (CIDd), a maioria (10 casos; 45,4%) foi diagnosticada entre 2 a 19 anos; 6 casos (27,3%), no mesmo período do seu respectivo CI; e 6 CN (3 dimorfo-tuberculóides e 3 tuberculóides) foram detectados durante a intervenção. Os CN evidenciaram elevada freqüência da positividade da RLEP-130 (50%) contrastante com a do anti-PGL-I (20%) e seu valor sérico médio (0,075) inferior. Entre os CI após a alta PQT-MB (>10 anos), houve inferioridade da freqüência da positividade (14,3%) e do valor sérico médio do anti-PGL-I (0,072), estabelecendo uma correlação negativa (p=0,038,r=-0,32) com o tempo decorrido. Valores similares da freqüência de positividade para RLEP-130 foram encontrados em: CI com alta por cura (26,7%), em períodos superiores (25%) a 10 anos de conclusão do tratamento e CID saudáveis (20,5%). A positividade para a RLEP-130 foi mais freqüente (p<0,001) comparada à da RLEP-372. CONCLUSÕES: A taxa de recidiva após PQT-MB 24 doses é baixa e prevalece alta percentagem de cura (91,3%). É racional considerar que o anti-PGL-I e a detecção do DNA bacilar por RLEP-130 devam ser analisados em conjunto com os demais achados clínicos e laboratoriais, ainda que nossos resultados possam ter diferenciado os grupos nas distintas condições: recidiva e alta por cura, e identificado doentes, contatos e famílias em risco. O seguimento e o monitoramento clínico e laboratorial por períodos prolongados incrementariam a detecção precoce de novos casos entre os comunicantes, desde que o intervalo para o diagnóstico foi longo para a maioria daqueles que adoeceram. Estas estratégias seriam potencialmente facilitadas em áreas com reduzida prevalência, e particularmente valiosas para a manutenção do controle na fase de pós-eliminação / Introduction: Multidrugtherapy (MDT-WHO) resulted in marked reduction of leprosy prevalence in Brazil, without impact on detection of new cases in Santa Catarina (SC) state in the post-elimination phase. Lepromatous cases with high bacilloscopic index have late relapse risk and can consist in the disease transmission focus. Objectives: The aim of this study was to evaluate the disease recurrence and microbiological resistance in lepromatous leprosy patients regularly treated with MDT-WHO/24 doses; specific immune response (cellular and humoral) and M. leprae DNA detection in index cases (IC) and their household contacts (HHC); and the findings face to the epidemiological and operational indicators of Santa Catarina state. Casuistic and methods: After consulting the Brazilian Information System of Notification (SINAN) database, 46 IC successfully diagnosed and treated between 1990 and 2000, and their 187 HHC from Joinville and Itajaí (SC) municipalities were selected. A dermatoneurological examination was performed, as well as the skin biopsies for histopathology and therapy resistance assay in mouse pads, skin smears for bacilloscopy, anti-PGL-I IgM serology, Mitsuda reaction, polymerase chain reaction for M. leprae DNA detection in nasal secretion based on 130bp and 372bp specific repetitive sequences (RLEP). Results: Cured IC (m=11.2±3 years) had mean age higher ((57.3±14.5 years old; p<0.05) than HHC, most of them were males (p=0.001) and Mitsuda negative (78.6%). In 4 relapsed cases (8.7%) the average anti-PGL-I serum levels (0.365), as well as frequency of positive antibody and RLEP-130 (75%; p=0.03) were consistently high and the mouse footpads assays resulted negative. Among 187 HHC, 22 (11.8%) became sick (sHHC), 10 cases (45.4%) were diagnosed between 2 and 19 years, being 6 cases (27.3%) in the same year of their IC; 6 new cases (3 borderline-tuberculoid and 3 tuberculoid) were detected during the study, with high RLEP-130 positive frequency (50%), as opposed to anti-PGL-I (20%) and low average serum levels (0.075). Among IC with more than 10 years of discharge, frequency of positive anti-PGL-I (14.3%) and average serum levels (0.072) were lower and had negative correlation (p=0.038, r= -0.32) with time after cure. Similar values of frequency of positivity for RLEP-130 were found: IC with discharge by cure (26.7%), when time interval was higher than 10 years (25%) and in healthy HHC (20.5%). RLEP-130 positivity was more frequent (p<0.001) than RLEP-372. Conclusions: Relapse rate after MDT-WHO 24 doses was low, with high cure rate (91.3%). Serology anti-PGL-I and M. leprae DNA detection by RLEP-130 must be analyzed together with other clinical and laboratory findings, though our results had differentiated groups in the following conditions: relapse and discharge by cure, patient identification, HHC and families at risk. Long term follow-up with laboratorial and clinical monitoring could lead to early detection of new cases among HHC, since the period between diagnoses was long for most sHHC. This strategy may be useful in areas with decreased prevalence, and of particular value for maintenance of disease control in the post-elimination phase

The bedrock under Griesgletscher is formed in a bowled-shaped cavity, an overdeepening. This is known to affect the flow of ice and subglacial water by causing inefficient drainage through the overdeepening. This report aims to, from field data, investigate what further consequences overdeepenings might have for subglacial drainage mechanisms and seasonal evolution. A field campaign was performed at Griesgletscher, Switzerland, during the summer 2017. Turbidity, discharge and electrical conductivity were monitored in a proglacial stream throughout the ablation season. 115 water samples were collected for calibration of the turbidity measurements and the relationship between the concentration of suspended sediments in the meltwater and the magnitude of discharge was investigated through simple and multiple linear regression. In addition, ten tracer experiments were conducted by injecting dye in moulins on the glacier tongue, and measuring the fluorescence of the water in the proglacial streams. The results obtained during the field campaign suggest that the main part of the overdeepened area at Griesgletscher is drained via a lateral channel passing around the overdeepening and that subglacial water from the overdeepening is drained at times of high discharge. The driving force for drainage of subglacial water from the overdeepening is suggested to be the gradient created from rising water pressure in the ice above the overdeepening. Results further suggest that subglacial drainage at a part of the adverse slope is inefficient and remains inefficient throughout the ablation season. This differs from the seasonal evolution normally seen at nonoverdeepened glaciers, in which an efficient, channelized system evolves during the course of the season. / Formen på berggrunden under en glaciär är avgörande för flödet av smältvatten och is. Det är vanligt att berggrunden under en glaciär utgör en skålformad fördjupning kallad överfördjupning. Det är sedan tidigare känt att flödet genom överfördjupningar är begränsat och forskning tyder på att smältvatten tenderar att välja kanaler över eller runt överfördjupningen i de fall då sådana finns tillgängliga. Många frågor kvarstår dock gällande vilken betydelse överfördjupningar har för det subglaciala dräneringssystemet och dess säsongsutveckling och antalet fältstudier som berör ämnet är få. Den här rapporten syftar till att, från fältdata, tillföra kunskap om överfördjupningars inverkan på subglacial dränering. Under sommaren 2017 utfördes en fältstudie på Griesgletscher, en överfördjupad glaciär belägen i de Schweiziska alperna. Turbiditet, elektrisk konduktivitet och flöde mättes i en av de proglaciära smältbäckarna. 115 vattenprover samlades in för bestämning av koncentrationen suspenderade sediment och användes för att omvandla mätningar av turbiditet till koncentration av suspenderade sediment. Relationen mellan transport av suspenderade sediment och flöde kunde sedan undersökas genom enkel och multipel linjär regression. I tillägg utfördes tio försök då ett fluorescerande spårämne injicerades i en moulin på glaciärtungan och fluorescensen mättes i smältbäckarna som avrinner från Griesgletscher. Utifrån resultaten kunde Griesgletschers dräneringssystem och dess utveckling kartläggas. De resultat som uppnåtts under fältstudien tyder på att det överfördjupade området av Griesgletscher i huvudsak dräneras via en sidokanal som passerar runt överfördjupningen samt att subglacialt vatten från överfördjupningen främst dräneras vid högt flöde. Drivkraften för dränering av subglacialt vattnet från överfördjupningen föreslås vara den gradient som skapas vid hög avrinning, då vattentrycket i isen ovanför överfördjupningen stiger. Resultatet från de regressionsanalyser som utförts visade sig vara representativt enbart för en del av det överfördjupade området. Tolkningar av resultaten tyder på att det subglaciala dräneringssystemet i detta område var ineffektivt att evakuera smältvatten och förblev ineffektivt under hela smältsäsongen. Detta skiljer sig från den säsongsmässiga utvecklingen som normalt kan ses på glaciärer utan överfördjupning, där ett effektivt kanaliserat dräneringssystem utvecklas under sommaren. Slutligen kunde graden av flöde fastställas som den viktigaste drivvariabeln för koncentrationen av suspenderade sediment i smältvattnet tillsammans med förändringen av flöde, tidigare koncentration av suspenderade sediment och nederbörd.

This paper seeks to conduct a first test of the explanatory potential of the poliheuristic theory of foreign policy decision-making in the context of small states. The case studied is the Swedish decision to contribute to the UN-sanctioned and NATO-led Operation Unified Protector in 2011. The paper conducts a theory-testing Causal Process Tracing (CPT) study drawing on a variety of different sources including news articles, parliamentary records, government bills, official statements and remarks made by key individuals, and secondary sources. The result of the analysis demonstrate the potential validity of the poliheuristic understanding of the decision- making process operating in a small state, but fall short of demonstrating actual validity. The final results are thus more akin to results typically found in a plausibility probe case study, and future research is deemed merited based on the potential validity found.

Ce travail de thèse concerne la modélisation du canal de propagation dans les milieux industriels. Dans ce contexte, le canal de propagation a un comportement différent de celui classiquement rencontré en indoor. Cela est dû à l'aménagement des bâtiments qui sont plus grand et ouverts ainsi qu'à la présence de machines, d'objets mobiles et d'autres matériaux métalliques rencontrés dans ces environnements. Ainsi, les modèles de canaux indoor existants ne sont plus valides. L'utilisation de modèles déterministes comme alternative est possible mais limitée en raison du temps de calcul qui en découle.Pour répondre à cette problématique, nous proposons un modèle hybride de canal s'inspirant d'une méthode à tracer de rayons 3D et du modèle WINNER. L'originalité de ce modèle repose sur son caractère hybride consistant, en prétraitement, à partitionner l'environnement en zones de visibilité ou de non-visibilité «faible» et «forte» sur des critères déterministes liés à la propagation des ondes. Un modèle statistique, type WINNER, reprenant le concept de cluster est ensuite "joué" au sein de chacune des ces zones reproduisant ainsi fidèlement l'évolution des paramètres caractéristiques des clusters identifiés. Nous avons validé notre modèle en le comparant d'abord à un modèle déterministe et ensuite à la mesure. Sa robustesse ainsi que celle de WINNER sont testées en les simulant dans trois environnements différents et en les comparant au modèle déterministe à tracer de rayons. / This thesis focuses on the modeling of the propagation channel in industrial environments. In this context, the propagation channel has a different behavior than typically encountered in indoor. This is due to the construction of buildings that are larger and open and the presence of machines, moving objects and metal materials encountered in these environments. Thus, the existing indoor channel models are not valid. Using deterministic models as an alternative is possible, but limited by the computing time.To address this problem, we propose a hybrid channel model for communications in industrial environments inspired by a ray tracing method and Winner model. The originality of this model is its hybrid nature consisting, in preprocessing, in partitionning the environment in areas of visibility or non-visibility «weak» and «strong» based on deterministic criteria related to the wave propagation. A statistical model, as WINNER , using the concept of cluster is then played in each of these areas and faithfully reproducing the evolution of the characteristic parameters of the identified clusters. We have validated our model by comparing it firstly to a deterministic model and then to measurement. Its robustness as well that of WINNER are tested by simulating them in three different environments and by comparing them with the deterministic model.

The Clement Ader Institute of Albi and the von Karman Institute follow<p>since numerous years works about IR radiometry with the aim to do quantitative<p>thermography (true temperature measurement without contact). These works allowed<p>to explore potentiality several spectral bands :8-12µm, 3-5µm and recently the<p>0.75-1.7µm band (near IR) with the help of CCD camera (Si) or VisGaAs camera.<p>Studies done in this specific domain have underlined some perturbations emphasized<p>at high-temperature (T>800◦C). This work has to deal in details with the<p>treatment of parameters playing a role in camera measurements. The first of these<p>parameters is the emissivity, this treatment is made in another thesis. The second<p>parameter affects particularly the hot spot location and the spatial distortion. This<p>perturbation comes from convective effect present around every hot objects.The<p>purpose of this thesis is to analyze this effect and to correct it. Indeed, when a hot<p>object is in a colder surrounding media, a temperature gradient is shaped around<p>the object and thus a refractive index gradient too. This phenomena brings inevitably<p>distortions of the spatial information received by the camera. The goal of this<p>work will be to estimate and correct error made on temperature and/or distortion<p>measurement by CCD or VisGaAs camera on hot object.We chose to focus our work<p>on the convective plume created by a hot horizontal disk. This study will be done<p>with an experimental and a numerical approach. For the numerical approach, a raytracing<p>code has been developed in order to obtain numerically the displacement<p>due to the light deviation occurring in the perturbation. The input data of the code<p>is the refractive index of the hot air present around the object. This refractive index,<p>depending on the wavelength, can be found directly from the temperature thanks<p>to the Gladstone-Dale law. The temperature is given by a CFD software such as<p>FLUENT. Experimentally, we will use the Background Oriented Schlieren (BOS)<p>method in order to retrieve the displacement. We can see that the displacement can<p>reach 4 pixels in the plume (corresponding to 1mm in this case). This perturbation<p>has been studied for several spectral bands (visible, near infrared, infrared). Finally,<p>some solutions of correction are given, like using the inverse Abel Transform<p>in order to retrieve from the 2D displacement, a 3D (axisymmetric) refractive index<p>field that we will implement in the ray-tracing code and consequently predict the<p>displacement for any kind of wavelength or distance (distance between the camera<p>and the object)./L’Institut Clément Ader Albi (ICAA) et l’Institut von Karman (IVK)<p>mènent depuis un certain nombre d’années des travaux sur la radiométrie IR dans<p>le but de faire de la thermographie quantitative (mesure de température vraie sans<p>contact). Ces travaux ont permis d’explorer les potentialités de plusieurs bandes<p>spectrales :8-12µm, 3-5µm et plus récemment la bande 0,75-1,7µm (proche IR) à<p>l’aide de caméras CCD (Si) ou VisGaAs. Les travaux effectués dans ce domaine<p>spectral ont permis de mettre en évidence un certains nombre de perturbations<p>renforcées par les hautes températures (T>800◦C). Cette thèse aborde de façon détaillée<p>le traitement d’un certain nombre de grandeurs d’influence liées à la mesure<p>de différents paramètres dans le domaine du proche IR mais également étendus aux<p>domaines du visible et de l’IR. La première de ces grandeurs est l’émissivité dont le<p>traitement a déjà été abordé par d’autres études. La seconde grandeur d’influence<p>touche plus particulièrement à la localisation des points chauds sur l’objet et la distorsion<p>du champ de température apportée par les effets convectifs présents autour<p>d’un objet à haute température, elle est le coeur de cette thèse. En effet, lorsqu’une<p>pièce chaude se trouve dans un milieu ambiant beaucoup plus froid, il se crée un<p>gradient de température et donc d’indice de réfraction autour de la pièce. Or les<p>caméras travaillant dans les différentes bandes spectrales vont être plus ou moins<p>sensibles à ces variations d’indices de réfraction du fait de la dépendance de l’indice<p>optique avec la longueur d’onde et de la résolution spatiale de la caméra utilisée.<p>Ce phénomène, appelé effet mirage, entraîne inévitablement une déformation des<p>informations spatiales reçues par la caméra. Le but de cette thèse a donc été d’estimer<p>et de proposer une première approche pour corriger l’erreur faite sur la mesure<p>de température et/ou de déformation faites par caméras sur des pièces chaudes.<p>La démarche générale du travail a donc été dans un premier temps de calculer le<p>champ de température autour de l’objet considéré en se ramenant d’abord à des cas<p>simplifiés. On en a déduit alors le champ de réfraction entraînant une « déformation<p>» de l’objet, en faisant le lien entre T et n. Cette étape correspond à l’approche<p>numérique de notre étude. Cette étape numérique a été réalisé à l’aide d’un outil de<p>lancer de rayons développée à l’ICA. L’approche expérimentale a consisté à l’utilisation<p>de méthodes telle que la BOS (Background Oriented Schlieren), la PIV,<p>la srtioscopie afin de déduire le champ de déplacements provoqué par le panache<p>convectif. Ces résultats ont été comparés à la méthode numérique et ceci pour différentes<p>longueurs d’ondes. Enfin, une stratégie de correction d’images perturbées<p>a été abordé à l’aide de méthodes telles que la transformée d’Abel inverse afin de<p>remonter au champ d’indice de réfraction 2D axisymétrique à partir d’une déformation<p>plane. / Doctorat en Sciences de l'ingénieur / info:eu-repo/semantics/nonPublished

[pt] O objetivo deste trabalho é desenvolver uma solução completa para o traçado de raios de cenas dinâmicas utilizando a GPU. Para que este algoritmo atinja desempenho interativo, é necessário utilizar uma estrutura espacial para reduzir os testes de interseção entre raios e triângulos da cena. Observa-se que, quando há movimento na cena, é necessário atualizar esta estrutura de aceleração, seja alterando-a parcialmente ou reconstruindo-a inteiramente. Adotamos a segunda estratégia por ser capaz de tratar o caso geral de movimento não-estruturado. Como a construção da estrutura deve ser feita da forma mais eficiente possível, escolhemos utilizar uma Grade Uniforme como foco de nossa pesquisa. Suas vantagens incluem um algoritmo de construção simples e um percurso de raios eficiente. Para explorar o poder de processamento em paralelo de uma GPU, é necessário manter os dados da cena e da estrutura de aceleração dentro da placa gráfica, evitando transferências custosas de memória entre a GPU e a CPU. Propomos neste trabalho uma técnica para construir uma grade uniforme inteiramente na GPU. Usando nosso método, é possível reconstruir toda a estrutura em poucos milissegundos, enquanto mantém-se a alta qualidade da grade obtida. Além disso, propomos uma implementaçoes do algoritmo de traçado de raios de forma a aproveitar o processamento em paralelo da GPU. Nosso procedimento é implementado inteiramente dentro da placa gráfica, onde há acesso direto para os dados dos triângulos da cena, bem como as informações da grade uniforme construída. Utilizando a solução proposta, somos capazes de obter taxas de visualização interativas mesmo para cenas com movimentos não-estruturados, incluindo texturas, sombras e até mesmo reflexões. / [en] We present a technique for ray tracing dynamic scenes using the GPU. In order to achieve interactive rendering rates, it is necessary to use a spatial structure to reduce the number of ray-triangle intersections performed. Whenever there is movement in the scene, this structure is entirely rebuilt. This way, it is possible to handle general unstructured motion. For this purpose, we have developed an algorithm for reconstructing Uniform Grids entirely inside the graphics hardware. In addition, we present ray-traversal and shading algorithms fully implemented on the GPU, including textures, shadows and reections. Combining these techniques, we demonstrate interactive ray tracing performance for dynamic scenes, even with unstructured motion and advanced shading effects.

Les travaux présentés dans ce mémoire portent sur l’étude et le développement d’une méthode de mesure d’épaisseur et de reconstruction 3D d’objets en verre creux, dans le cadre d’une application de contrôle industriel sur ligne de production. Les techniques de mesure d’épaisseur de parois transparentes sont actuellement très limitées par leur faible tolérance angulaire sur l’orientation des surfaces. Elles nécessitent une rotation de l’article devant le capteur, donc une manutention contraignante. De plus, les surfaces considérées étant spéculaires, les techniques de numérisation standard ne fonctionnent pas. Nous avons développé deux méthodes basées sur l’observation des réflexions d’une source diffuse ponctuelle sur les deux surfaces de la paroi de l’objet. Une observation simple à l’aide d’une caméra implique une ambiguïté entre l’orientation et la position de chacune des deux surfaces. Notre première méthode consiste à utiliser des informations de polarisation pour lever cette ambiguïté. Des résultats expérimentaux, obtenus avec un ou plusieurs points source témoignent de la faisabilité de la mesure. La seconde méthode consiste à utiliser une seconde caméra, associée à un dispositif optique spécifique. Ceci permet de trianguler directement les points d’incidence sur les deux surfaces. De plus, la source peut être étendue verticalement, de manière à augmenter la taille de la zone inspectée. Les résultats obtenus avec cette seconde technique attestent de la possibilité d’une mise en oeuvre industrielle. / The work presented in this thesis focuses on the study and development of a method for measuringthe thickness and reconstructing the 3D shape of glass containers, within the frame of an industrialon-line quality monitoring. Measuring thickness techniques of transparent objects are currently limited by their lack of toleranceregarding the surface orientation at a given reflection point. A rotation of the article in front ofthe sensor is thus required, which adds complexity to the mechanical handling of the item beingmeasured. Moreover, conventional scanning techniques do not work well on specular surfaces.We developed two methods based on the observation of the reflections by both surfaces of the wallof a given item of a diffuse, point light source. Imaging these reflections with a plain camera leavesan ambiguity between the orientation and position of each surface. Our first approach uses light polarization to remove this ambiguity. Experimental results, obtainedwith one or several point sources have demonstrated the feasibility of the measure. The second approach uses a second camera, associated with a specific optical design, whichenables a mere triangulation of the incident points on the two surfaces. Moreover, the light sourcecan be vertically extended to increase the size of the inspected area. The results obtained with thissecond technique show that industrial implementation is feasible.

Les répertoires iconographiques rupestres du Sahara central ont pâti du contexte historique et humain de leur découverte, puis de leur gestion tant scientifique que patrimoniale. Leur profusion et les difficultés du terrain saharien ont été et sont toujours des facteurs pénalisants, mais la méconnaissance dont souffrent encore ces productions des cultures préhistoriques, plus d’un siècle et demi après leur révélation, incombe clairement aux acteurs passés et présents de leur étude. Même si nous admirons ces images pour leurs qualités esthétiques, ce sont des vestiges archéologiques à part entière, devant être étudiés comme tels. De nos jours, les outils numériques et informatiques permettent, d’une part, d’améliorer quantitativement et qualitativement leur documentation grâce aux logiciels d’amélioration d’image utilisés pour l’enregistrement et le relevé ; d’autre part, d’interroger plus objectivement ces œuvres par le biais d’autres outils et de méthodes performantes d’analyse des données, au travers desquels leur différenciation stylistique peut être mise en évidence. L’application des démarches développées dans cette recherche à quelques parois ornées du site de Séfar, lieu emblématique de la Tasīli-n-Ăjjer (Algérie), montre que l’usage de ces outils, sous-tendu par des bases théoriques et méthodologiques rigoureuses, permet de proposer une nouvelle approche de ces images rupestres, plus conforme à leur nature de données scientifiques. / The rock images of the central Sahara have suffered from the historical and human context of their discovery and then from both their scientific management and their heritage management. Their abundance and the difficulties of the Saharan field have been and are still negative factors, but the lack of knowledge from which these productions of prehistoric cultures are still suffering more than a century and a half after their discovery, clearly falls to those studying them in the past and present. Even if we admire these images for their aesthetic qualities, these form an integral part of the archaeological remains and should be studied as such. Nowadays, the digital and computerised tools enable us, firstly, to quantitatively and qualitatively improve their documentation thanks to image enhancement software used for the recording and the tracing; secondly, to more objectively investigate these artefacts by using other effective data-analysis tools and methods thanks to which their stylistic differences can be highlighted. The application of the processes developed in this research to several adorned walls at the Sefar site, an emblematic area of the Tasīli-n-Ăjjer (Algeria), shows that the use of these tools, supported by rigorous theoretical and methodological basis, allows us to offer a new approach to these rock images, more appropriate to their characteristic of being scientific data.