Abstract Background: The discovery of brown adipose tissue (BAT) in adults has sparked interest in its role as a therapeutic target in metabolic disorders. Preliminary studies have indicated that infrared thermography may be a promising way to quantify this thermogenic organ, which in humans is located primarily in the supraclavicular area and is activated by cold. However, ways to standardize infrared thermography methodology and to ensure measurements are reproducible have not been established. Objective: This study aims to establish a standardized and reproducible protocol to measure a thermal response to cold in the supraclavicular area. Method: In phase 1 of the study, thermal images of the supraclavicular area were taken on 3 occasions in 28 healthy adult males with mean age 23.95 ±5.87 years and mean BMI 25.20 ±3.93 kg/m2 who demonstrated a 100kcal/d increase in energy expenditure when exposed to 12ºC; a temperature known to increase BAT activity without shivering. During the first and second visits, participants were acclimated for 1 hour at 32ºC and room temperature (20-23ºC) respectively, followed by a 1 hour period of torso cold exposure at 12ºC using a cooling blanket. The third visit consisted of taking thermal images at room temperature over two hours. Body composition was measured with DEXA scanner. In phase 2, 3 trials of 32ºC acclimation followed by 12ºC cold exposure (32ºC-cold) were studied in 14 healthy adult males (mean age 20.93 ±2.4 years and mean BMI 23.55 ±3.15 kg/m2) for repeatability. The outdoor temperature on the morning of each visit was recorded from the website http://climate.weather.gc.ca in both phases. Results: In phase 1 the supraclavicular temperature stabilized after 45 minutes of acclimation at 32ºC and then rose abruptly with cooling, plateauing at 10 minutes. The change in supraclavicular temperature in response to cooling was greater after 32ºC compared to room temperature acclimation (0.22 ±0.19 vs 0.13±0.17ºC, p=0.053). There was no relation between outdoor temperature on the morning of the visit and the 32°C-cold thermal response (r=-0.18, p=0.14). The 32ºC-cold thermal response did not correlate with cervical and supraclavicular fat in 25 young males with BMI between 19.3-32.3 kg/m2 (r=-0.26, p=0.21). In phase 2, the thermal response after 32ºC acclimation was reproducible [intraclass correlation coefficient of 0.69 (0.14-0.72)]. Conclusion: Acclimation at 32ºC produces a greater and earlier response to cold in the supraclavicular area than room temperature acclimation. The thermal response after 32ºC acclimation is reproducible and unlikely to be affected by outdoor temperature and subcutaneous fat in the neck. These data suggest that the use of infrared thermography using the 32ºC-cold protocol may be effective for detecting the metabolic activity of brown adipose tissue. / Thesis / Master of Science (MSc)

The abilities of powerful regression and multi-type data processing allow deep learning to effectively and accurately complete multi-tasks, which is the need of civil engineering. More cases showed that deep learning has become a greatly powerful and increasingly popular tool for civil engineering. Based on these, this dissertation developed deep learning studies for the condition assessment of civil infrastructure and materials. This dissertation included five main works: (1) Deep learning and infrared thermography for asphalt pavement crack severity classification. This work focused on longitudinal or transverse cracking. This work first built a dataset with four severity levels (no, low-severity, medium-severity, and high-severity) and three image types (visible, infrared, and fusion). Then this work applied the convolutional neural network (CNN) to classify the crack severity based on two strategies deep learning from scratch and transfer learning). This work also investigated the effect of image types on the accuracy of these two strategies and on the classification of different severity levels. (2) Asphalt pavement crack detection based on convolutional neural network and infrared thermography. This work first built an open dataset with three image types (visible, infrared, and fusion) and different conditions (single, multi, thin, and thick cracks; clean, rough, light, and dark backgrounds) and periods (morning, noon, and dusk). Then this work evaluated the performance of the CNN model based on the accuracy and complexity (computational and model). (3) An artificial neural network model on tensile behavior of hybrid steel-PVA fiber reinforced concrete containing fly ash and slag powder. This work considered a total of 23 factors for predicting the tensile behavior of hybrid fiber reinforced concrete (HFRC), including fibers' characteristics, mechanical properties of plain concrete, and concrete composition. Then this work compared the performance of the artificial neural network (ANN) method and the traditional equation-based method in terms of predicting the tensile stress, tensile strength, and strain corresponding to tensile strength. (4) Deep transfer learning-based vehicle classification by asphalt pavement vibration. This work first applied the pavement vibration IoT monitoring system to collect raw vibration signals and performed the wavelet transform to obtain denoised vibration signals. Then this work represented the vibration signals in two different ways, including the time-domain graph and the time-frequency graph. Finally, this work proposed two deep transfer learning-based vehicle classification methods according to these two representations of vibration signals. (5) Physical-informed long short-term memory (PI-LSTM) network for data-driven structural response modeling. This work first applied the single-degree-of-freedom (SDOF) system to investigate the performance of the proposed PI-LSTM network compared with the existing methods. Then this work further investigated and validated the proposed PI-LSTM network in terms of the experimental results of one six-story building and the numerical simulation results of another six-story building. / Doctor of Philosophy / With the development of technologies, deep learning has been applied to numerous fields to improve accuracy and efficiency. More work shows that deep learning has become a greatly powerful and increasingly popular tool for civil engineering. Since civil infrastructure and materials play a dominant role in civil engineering, this dissertation applied deep learning to the condition assessment of civil infrastructure and materials. Deep learning methods were applied to detect cracks in asphalt pavements. The mechanical properties of fiber reinforced concrete were investigated by deep learning methods. Based on the asphalt pavement vibration, the type of vehicles was classified by deep learning methods. Deep learning methods were also used to investigate the structural response.

The disintegration of a liquid jet emerging from a nozzle has been under investigation for several decades. A direct consequence of the liquid jet disintegration process is droplet formation. The breakup of a liquid jet into discrete droplets can be brought about by the use of a diverse forcing mechanism. Cavitation has been thought to assist the atomization process. Previous experimental studies, however, have dealt with cavitation as a secondary phenomenon assisting the primary atomization mechanism. In this dissertation, the role of the energy created by the collapse of cavitation bubbles, together with the liquid pressure perturbation is explicitly configured as a principal mechanism for the disintegration of the liquid jet. A prototype of an atomizer that uses this concept as a primary atomization mechanism was developed and experimentally tested using water as working fluid. The atomizer fabrication process and the experimental characterization results are presented. The parameters tested include liquid injection pressure, ultrasonic horn tip frequency, and the liquid flow rate. The experimental results obtained demonstrate improvement in the atomization of water. To fully characterize the new atomizer, a novel infrared thermography-based technique for the characterization and visualization of liquid sprays was developed. The technique was tested on the new atomizer and two automotive paint applicators. The technique uses an infrared thermography-based measurement in which a uniformly heated background acts as a thermal radiation source, and an infrared camera as the receiver. The infrared energy emitted by the source in traveling through the spray is attenuated by the presence of the droplets. The infrared intensity is captured by the receiver showing the attenuation in the image as a result of the presence of the spray. The captured thermal image is used to study detailed macroscopic features of the spray flow field and the evolution of the droplets as they are transferred from the applicator to the target surface. In addition, the thermal image is post-processed using theoretical and empirical equations to extract information from which the liquid volume fraction and number density within the spray are estimated.

L'ébullition est présente dans la vie de tous les jours et elle a été par conséquent le sujet de beaucoup d'études, mais pour la plupart en régimes stationnaires. Néanmoins, l'intérêt de connaître les caractéristiques de l'ébullition transitoire est aussi important notamment pour la prévention des accidents nucléaires majeurs. C'est justement dans l'optique de mieux comprendre les phénomènes d'ébullition qui se produisent lors d'un RIA (Accident d'Insertion de Réactivité) que cette thèse a été financée par l'IRSN. Le RIA est un accident qui peut résulter d'une défaillance du mécanisme de la grappe contrôlant la réaction nucléaire. La réaction s'emballe pendant quelques dizaines de millisecondes (pulse de puissance) provoquant une augmentation rapide de la température du crayon de combustible et donc l'évaporation du liquide qui l'entoure. Des tests ont été faits par le passé soit sur des crayons de combustibles, soit sur des tubes chauffés ayant les mêmes dimensions qu'un crayon, afin d'améliorer la connaissance de ce phénomène. Par contre, les mesures étaient entachées d'incertitudes importantes, dues à des techniques de mesure non appropriées à des phénomènes si rapides. L'objectif de ce travail a été de concevoir et mettre en place une expérience capable de simuler un RIA à petite échelle, pour mieux comprendre les caractéristiques de l'ébullition lorsque la paroi monte en température très rapidement. De plus, ce dispositif expérimental devait être apte à étudier des montées en température moins violentes pour améliorer la connaissance de l'ébullition transitoire en général. Cette expérience a été conçue à l'Institut de Mécanique des Fluides de Toulouse. Elle est constituée d'une feuille métallique d'acier de 50µm d'épaisseur, formée en demi cylindre (8mm de diamètre et 200mm de longueur) et chauffée par effet Joule. Elle est entourée par du fluide réfrigérant HFE7000, qui permet de travailler en similitude par rapport au cas réel en eau. Le fluide est confiné par un deuxième demi cylindre en verre, ayant 34mm de diamètre. Les expériences peuvent être en vase ou avec écoulement, écoulement qui a été caractérisé par des mesures PIV. Plusieurs débits peuvent donc être employés et le sous-refroidissement du liquide est aussi ajustable. L'emploi d'une alimentation pilotable et très flexible permet d'obtenir des chauffages du métal jusqu'à 2500K/s, mais aussi des montées en température plus faibles, pour tracer des courbes d'ébullition stationnaires ou faiblement transitoires. La température de la paroi est mesurée grâce à une caméra infrarouge, couplée à des visualisations rapides et à des mesures de pression et température dans le liquide. / Boiling phenomena can be found in the everyday life, thus a lot of studies are devoted to them, especially in steady state conditions. Transient boiling is less known but still interesting as it is involved in the nuclear safety prevention. In this context, the present work was supported by the French Institute of Nuclear Safety (IRSN). In fact, the IRSN wanted to clarify what happens during a Reactivity-initiated Accident (RIA). This accident occurs when the bars that control the nuclear reactions break down and a high power peak is passed from the nuclear fuel bar to the surrounding fluid. The temperature of the nuclear fuel bar wall increases and the fluid vaporises instantaneously. Previous studies on a fuel bar or on a metal tube heated by Joule effect were done in the past in order to understand the rapid boiling phenomena during a RIA. However, the measurements were not really accurate because the measurement techniques were not able to follow rapid phenomena. The main goal of this work was to create an experimental facility able to simulate the RIA boiling conditions but at small scale in order to better understand the boiling characteristics when the heated-wall temperature increases rapidly. Moreover, the experimental set-up was meant to be able to produce less-rapid transients as well, in order to give information on transient boiling in general. The facility was built at the Fluid-Mechanics Institute of Toulouse. The core consists of a metal half-cylinder heated by Joule effect, placed in a half-annulus section. The inner half cylinder is made of a 50 microns thick stainless steel foil. Its diameter is 8mm, and its length 200mm. The outer part is a 34mm internal diameter glass half cylinder. The semi-annular section is filled with a coolant, named HFE7000. The configuration allows to work in similarity conditions. The heated part can be place inside a loop in order to study the flow effect. The fluid temperature influence is taken into account as well. A flexible power supply that can generate a free-shape signal, allows to get to a wall-temperature increase rate up to 2500 K/s but also to obtain lower rates, which permits to study weaker transients and steady state conditions. The thermal measurements are realised by means of an infra-red camera and a high-speed camera is employed in order to see the boiling phenomena at the same time. From the voltage and current measurements the heat flux that is passed to the fluid is known.

Pelo fato da face ser uma estrutura que envolve multissistemas, ela é considerada anatomica e funcionalmente muito complexa em sua avaliação clínica. Muitos diagnósticos de doenças que afetam a região facial são realizados por meio de métodos de imagem, entretanto, as técnicas convencionais são incompletas, pois apresentam um déficit de informações em relação aos aspectos funcionais, relativos à microcirculação regional e o sistema nervoso autônomo. Para documentação objetiva destas alterações, a termografia por imagem infravermelha tem sido proposta como método auxiliar diagnóstico. O objetivo deste estudo foi identificar, mapear e quantificar pontos de referência baseados na termoanatomia da face, a partir de gradientes térmicos, em uma população adulta. A amostra constituiu-se por 161 voluntários, ambos os gêneros, entre 26 e 84 anos (63 ±15 anos). Foram identificados 28 pontos de referência termoanatômicos fixos em 94,6% da amostra. A temperatura média desses pontos de referência termoanatômicos variou entre 33,3°C a 35,6°C, sendo que o melhor valor de corte para sua correta identificação na face foi acima de 34,5ºC. Não houve diferença estatística quando os comparou entre as hemifaces correspondentes e nem quanto à idade dos voluntários (?T = 0,11°C), exceto entre os gêneros e grupo racial (p<0,05). Concluiu-se que a face apresenta pontos termoanatômicos fixos e confiáveis em adultos, a qual a termografia tem o potencial de auxiliar na identificação de alterações a partir destes pontos. / Anatomical and functional information of the structures of interest are the basis for interpretation of imaging exams. For recognizing diseases it is necessary to understand and identify the anatomical structures and its variations. Diagnosis of some facial diseases may be done through medical imaging techniques; however, the most conventional methods present lack of information regarding physiological and functional aspects such as microcirculation and autonomous nervous system. The aim of this study was to identify, map and quantify the facial thermal gradients (thermo-anatomical points) in healthy adult population. A hundred sixty-one volunteers of both gender and mean age of 63 ±15 years were included. The results showed that twenty eight thermo-anatomical points have been identified in 94.6% of the sample, with a significant difference of the temperature in gender and racial groups (p<0.05). Moreover, there was no statistically significant difference between ages and sides in healthy individuals (?T = 0.11°C). The average temperature of the points ranged from 33.3°C to 35.6°C. When the maximum temperature reached to detect a facial thermal gradient is higher than 34.5°C, the thermo-anatomical point is meant to be correctly delineated. For detection of functional changes from thermoanatomical points on the face, thermography may be used as an auxiliary diagnostic imaging method.

This thesis experimentally investigates the phenomenon of ingress through gas turbine rim seals. The work focuses on developing experimental and numerical techniques for measuring the required sealing flow levels to purge the wheel-space against ingress and the effect of externally-induced ingress on the surface temperature as well as heat transfer to the rotor. Ingress is driven by a pressure difference between the mainstream annulus and wheel-space cavity resulting from the asymmetric external pressure profile in the annulus and/or the rotation of fluid in the rotor-stator wheel-space cavity. It can be prevented by pressurising the wheel-space through the supply of sealant flow. The University of Bath had measured and shown, for the first time, the thermal effects of ingress on the rotor in the wheel-space for a datum seal (axial-clearance seal) using thermo-chromic liquid crystal. However, as the previously used experimental technique with thermo-chromic liquid crystal was prone to large uncertainties, a non-intrusive temperature measurement technique using an infrared (IR) temperature sensor was developed. The new technique was successfully applied to the Bath one-stage gas turbine test facility and provided a full temperature history of the rotor surface in a transient heat transfer experiment. Moreover, a data analysis method appropriate for transient experiments using the IR temperature measurement technique was developed. The method was used to accurately calculate the heat transfer coefficient and the adiabatic surface temperature based on the full temperature history. A series of numerical experiments was carried out to develop the analysis method and the results from the numerical experiments were used to design new heat transfer experiments for both the 1 and 1.5-stage ingestion rigs of the University of Bath. Gas concentration measurements were made on the stator of the Bath one-stage gas turbine test rig to determine the variation of sealing effectiveness with sealant flow rate for four different seal geometries at design operational conditions. The IR temperature measurement technique was used to determine the effect of ingress on the heat transfer coefficient and the adiabatic wall temperature on the rotor of the ingestion test facility. Concurrent gas concentration measurements were made on the stator to compare the effects of ingress on the two discs (stator and rotor). Comparison between the adiabatic effectiveness on the rotor and the concentration effectiveness on the stator showed that the rotor was protected against the effects of ingress relative to the stator. The sealing air, which was drawn into the rotor boundary layer from the source region, thermally buffered the rotor against the ingested fluid in the core. Subsequently, a thermal buffer ratio hypothesis was developed and shown to be in good agreement with the experimental data. A previously published orifice model was modified so that the sealing effectiveness determined from the concentration measurements in a rig could be used to determine the effectiveness based on pressure measurements in an engine. There was good agreement between the effectiveness acquired from pressure measurement determined using the theoretical model and the sealing effectiveness determined from concentration measurements. It was also shown how parameters obtained from measurements of pressure and concentration in a rig could be used to calculate the sealing effectiveness in an engine.

Defects in Aluminum, Stainless steel and galvanized steel sheets are studied in reflection mode infrared thermography. The effect of material properties, surface finish, heating intensity, heater emission spectra, pixel size and defect size are studied. Contrast is governed by heat quality, emissivity and defect geometry—which follows a logarithmic trend. The diameter detected via infrared thermography is found to be at least 30% larger than the measured diameter and sub-pixel defects can be detected. The use of gradient and Laplacian of temperature is introduced as a means of increasing defect contrast and mitigating heater variation.

Residual feed intake (RFI) is a recognized measure of biological efficiency in beef cattle. However, RFI determination is expensive, time consuming and not well studied in animals fed forage-based diets. The objectives of this experiment were to investigate infrared thermography (IRT) as method for determining RFI ranking in yearling beef bulls, and to evaluate the effect of diet type on RFI repeatability in consecutive feeding periods. No significant correlations (P > 0.05) were observed between eye or cheek surface temperatures measured using handheld or within-pen stationary infrared camera systems with RFI. Reranking was observed for RFI in all diet treatments, however significant repeatability estimates occurred for the forage and grain diet treatments (r=0.58 and 0.64 respectively; P < 0.01) but not the diet switch treatment (r=0.24; P > 0.05). Extreme cold temperatures experienced in Western Canada influence IRT measurements and energy partitioning, indicating the need for standardized performance testing procedures. / February 2016

Background: The Global Postural Re-education (GPR) method seems to be an effective method for the treatment of musculoskeletal disorders. However, no study has examined its effects on a group of musicians with musculoskeletal pain. Additionally, infrared thermography (IRT) has been widely used in the field of medicine as a monitoring a diagnostic tool, which can provide empirical data about the effectiveness of the GPR method. Objective: To examine the effects of GPR using scales and questionnaires and IRT. Methods: This thesis enclosed two articles. In the first, musicians with (N=6) and without (N=6) musculoskeletal pain underwent a thermographic evaluation to examine the skin temperature asymmetry (STA) between both groups as a representation of musculoskeletal pain. The second article was a randomized control trial and examined the effects of the GPR on a group of 13 participants with musculoskeletal pain (7 experimental, 6 controls) after receiving 8 sessions of GPR. Pain, disability and quality of life data was collected using the VAS, MPIIQM, SF-36. A thermographic assessment was also conducted to examine the correlation between pain severity and degree of STA, and changes in contralateral skin asymmetries before and after the intervention. Results: The first article revealed no significant differences in STA between groups. Results from the second article showed statistical significant improvements in pain, disability and quality of life on the group who received the GPR intervention. Results from the IRT analysis were not significant. Conclusion: Results from the self-report questionnaires suggest that GPR is effective in treating musculoskeletal pain in the targeted group of musicians. Pain changes did not correlate with thermal outcomes. Further studies are needed to confirm the results obtained with IRT.

The research work focuses on implementing intelligent measurement and diagnostic techniques for the non-destructive evaluation (NDE) of aircraft carbon composites. The outcome of this research work developed reliable and faster techniques to aid in the rapid assessment of defects in anisotropic carbon composites by applying ultrasonic and infrared thermography NDE methods. To fulfill the requirement of the intelligent non-destructive evaluation methods, this research is divided into four sub-researches: fuzzy logic based delamination detection, super-resolution image reconstruction for ultrasonic C-scan, ultrasonic 3D reconstruction, and polynomial fitting techniques for infrared thermography inspection. These researches focus on the improvement and optimization of current ultrasonic testing and infrared thermography inspection. They are independent but interrelated component, and they all serve the same goal which is to interpret data correctly and provide detailed information about the region of interests (ROI) for intelligent non-destructive measurement and evaluation. Details of these researches are presented in Chapter 2, 3, 4, and 5 respectively. For the ultrasonic testing, a fuzzy inference classifier will be used to generate the rule base and knowledge base for different kinds of defects in composites. It will automatically manage large amounts of signal data sets and extract the important information. Data features and NDE expert knowledge are seamlessly combined to provide the best possible diagnosis of the potential defects and problems. As a result, the outcome of this research work will help ensure the integrity and reliability of carbon composites. The C-scan image resolution of ultrasonic testing system was improved by applying super-resolution algorithms to overcome the inherent resolution limitations of the existing ultrasonic system. It greatly improves the image quality and allows for more detailed inspection of the ROI with high resolution, making defect evaluation easier and more accurate. The ultrasonic 3D reconstruction technique will be able to provide NDE inspectors with more detailed information on defect depth, volume, and 3D structure, as well as help them make quick, accurate, and reliable decisions. For the IR inspection, the thermography methods based on the thermal contrast are strongly affected by non-uniform heating which due to the heat source alignment and specimen thickness variation. The proposed polynomial curve fitting and surface fitting techniques were applied to eliminate the non-uniform heating effect by subtracting the estimated non-uniform heating pattern from the corrupted IR images. Mainly, aircraft composite material: carbon fiber reinforced polymer (CFRP) panels will be considered for this research work. Based on the preliminary study, delamination defects due to impact damage and foreign object inclusions artificially embedded in CFRP panels were successfully detected by immersion ultrasonic testing (UT) and IRT inspection. Therefore, the next step will be in improving the detection algorithm and developing an intelligent quality inspection technique for NDE testing. Powered with multiple image processing techniques and mathematical algorithms, the research result will provide high resolution images and detailed information about defect areas. In addition, it will also capable of identifying the type, shape, size, and the distribution of defect.