111 |
Sobreposição de imagens de termografia e ressonância magnética: uma nova modalidade de imagem médica tridimensionalSanches, Ionildo José 2010 October 1914 (has links)
A termografia por imagem infravermelha é uma técnica para diagnóstico não-invasivo que permite a avaliação e quantificação de mudanças de temperatura na superfície da pele. Uma vez que as variações de temperatura na superfície do corpo humano são contínuas, é difícil a associação de detalhes anatômicos bem como, associar as variações de temperatura da pele com as estruturas ou órgãos internos. Apesar de fornecer informações significativas para o diagnóstico médico, esta técnica não permite avaliar detalhes anatômicos da região estudada. Este trabalho apresenta uma nova metodologia para fusão de imagens de ressonância magnética 3D (ou tomografia computadorizada) e imagens infravermelhas (2D). O resultado é uma imagem termográfica 3D que pode auxiliar no diagnóstico médico de certas doenças, tais como insuficiência vascular periférica, pé-diabético, tumores de mama e tireóide, sinusite, cefaléia, doenças reumáticas e outros processos inflamatórios, isto é, doenças que podem ser diagnosticadas pelo mapeamento térmico da superfície do corpo. No desenvolvimento deste trabalho pluridisciplinar de Engenharia Biomédica, foram utilizados conhecimentos de Processamento Digital de Imagens, Computação Gráfica, Visão Computacional, Programação de Computadores, Radiologia, Imageamento Médico e Termografia, interagindo com áreas da Medicina, da Ciência da Computação e da Engenharia Elétrica. Em um primeiro experimento, são adquiridas as imagens por ressonância magnética (MRI) da região de interesse. Em um segundo experimento, adquire-se um conjunto de imagens térmicas. Gera-se projeções 2D a partir dos planos tomográficos na mesma direção/orientação do conjunto de imagens tomográficas. Em seguida realiza-se o registro desses dois conjuntos de imagens. Após o registro, as imagens térmicas são combinadas e projetadas sobre o modelo 3D das imagens de MRI ou CT. O resultado é uma imagem que combina informação de duas modalidades de imagens médicas diferentes. Os excelentes resultados obtidos com essa nova metodologia demonstram que ela pode auxiliar no diagnóstico médico. / Infrared (IR) imaging is a noninvasive diagnostic technique that allows the examiner to evaluate and quantify changes on skin surface temperature. Since changes over the human body surface temperature are continuous, it is difficult to associate anatomical details, as well to associate skin temperature with internal structures or organs. Though providing significant information to clinical diagnosis, this technique does not allow evaluating anatomical details of the imaged region. In this work it is presented a new image fusion method for combining 3D Magnetic Resonance Images (or Computed Tomography) and 2D thermal images. The result is a 3D thermographic imaging that can improve medical diagnosis of certain human diseases, such as peripheral vascular disease, diabetic foot, breast and thyroid tumors,sinusitis, cephalalgia, rheumatic disorder and others inflammations process, i.e., diseases that can be diagnosis through the body surface thermal mapping. In the development of this interdisciplinary work of Biomedical Engineering, the knowledge of Digital Image Processing, Computer Graphics, Machine Vision, Computer Programming, Radiology, Medical Imaging and Thermography, was necessary, interacting with areas of Medicine,Computer Science and Electrical Engineering. First, magnetic resonance images (MRI) or computed tomography (CT) from region of interest are acquired. In a second experiment, a set of IR images is acquired. From the MRI or CT data, 2D projections are generated in the same direction/orientation of the tomographic planes. Then, the generated projections and the IR images are registered. Next, the registered IR images set is combined and projected over the 3D-MRI or 3D-CT. The result is a fused image that combines the information contents from two different medical images modalities. The excellent results obtained with the new methodology demonstrate that it can aid medical diagnosis.
|
112 |
Micro-concentrateurs de lumière à base de métamatériaux polymères pour la détection infrarouge / Wavelength-scale light concentrators made of polymer metamaterials for infrared applicationsMoughames, Johnny 13 July 2016 (has links)
Dans cette thèse, nous développons des micro-concentrateurs de lumière en vue d’applications dans l’infrarouge. Les optiques, plates, de dimensions comparables à la longueur d’onde, sont formées d’une couche micronique de métamatériau, constituée de polymère structuré par des trous d’air. Une focalisation dans la zone de Fresnel des structures est recherchée en réalisant un gradient d’indice, obtenu en variant le diamètre des inclusions d’air (de λ//20 à λ/8). Des simulations électromagnétiques sont d’abord effectuées pour valider un design. La fabrication de ces concentrateurs repose sur la structuration d’une couche de photorésine à l’aide d’une technique de lithographie optique 3D à deux photons. Les mesures obtenues par l’imagerie infrarouge des structures montrent l’obtention d’une focalisation en accord avec les simulations pour un volume de concentrateur de 1,5 λ³, en dépit de l’absorption résiduelle de la résine choisie. Les structures proposées sont invariantes dans la direction axiale et peu épaisses, et peuvent donc être transférées dans d’autres types de matériaux ayant des indices de réfraction plus importants, comme le silicium. Les structures proposées peuvent également servir de plateforme pour réaliser un confinement sub-longueur d’onde par l’ajout d’antennes plasmoniques / In this thesis, we develop flat light concentrators for infrared applications. The structures have dimensions comparable with the wavelength and are made of a metamaterial layer (few microns) made of polymer with air holes inclusions. Light focusing in the Fresnel zone of the structures is achieved by a gradient index profile obained by chirping the holes diameter (from λ//20 to λ/8). Electromagnetic simulations are first performed to validate a design. The fabrication of these concentrators is then made by direct laser writing using a 3D two-photon lithography technique. Infrared imaging of the structures reveals a clear focusing of the infrared light for concentrators volume as small as 1,5 λ³, in agreement with the electromagnetic simulations. Considering that the metamaterial concentrator slabs are invariant in the axial direction and not too thick, similar structures can be transferred in transparent substrates such as silicon using deep reactive ion etching. A subwavelength light confinement can also be exploited by adding plasmonic antenna on the top surface of the flat concentrators
|
113 |
Analysis ex vivo biomechanics of two methods of osteosynthesis of pelvis in dogs / Thermographic printing in the detection of lameness in dogsGarcia, érika Fernanda Villamayor 20 December 2013 (has links)
Coordenação de Aperfeiçoamento de Pessoal de Nível Superior / Thermography is a non-invasive imaging diagnostic technique that records thermal patterns generated by infrared emission of heat from the body surface based on the local cutaneous perfusion. These thermal patterns are converted and displayed on a color map. In horses, since the 60s have been used for the detection of various orthopedic injuries and, in recent years studies in dogs have been published. Locomotion is the process by which the animal moves from one geographic location to another. The study of locomotion is essential to establish methods for treating disorders of the locomotor apparatus. Examination of the gait is the first step for the diagnostic in the evaluation of a patient with claudication. The aim of this study was to establish the thermal patterns of the paw prints of dogs with lameness compared to healthy dogs. It was verified also if there are differences in the temperature of paw print of the lame limb versus normal limb of normal dog and weight bearing limb of lame dog versus normal limb of normal dog. It was used 14 dogs with unilateral pelvic limb lameness due to rupture of the cranial cruciate ligament (lame limb versus weight bearing limb) and 14 healthy dogs (right limb versus left limb). Thermographic images of the paw print were made after the dog remained in an orthostatic position on foam mat for 30 seconds. It was analyzed average temperatures and thermal patterns. Gait analysis was performed on a force platform in the same position to measure the peak vertical force (PVF). It was calculated mean, standard deviation, percentage of average, and asymmetry index for this variable of the gait. There were no significant differences in the temperature of the paw print between groups. Analysis of the pattern of thermal images showed 80% success in differentiating lame limb versus weight bearing limb (abnormal group) and 100 % in the identification of the same thermal pattern between right limb compared to the left limb (normal group). The mean of the PVF showed 10.0 % difference in normal dogs and 72.4 % in abnormal dogs. Analysis of asymmetry index showed 5 % in the normal group and 36.2 % in the abnormal group. It is conclude that the thermal paw print can be successfully used as an additional tool for the detection of the pelvic limb lameness in dogs. / Termografia é uma técnica não invasiva de diagnósticos por imagem, que registra padrões térmicos gerados pela emissão infravermelha de calor da superfície corporal baseado na perfusão cutânea local. Estes padrões térmicos são convertidos e visualizados em um mapa de cores. Em equinos, desde os anos 60 têm sido utilizada para a detecção de várias lesões ortopédicas e, nos últimos anos, estudos em cães têm sido publicados. A locomoção é o processo pelo qual o animal se move de uma posição geográfica para outra. O estudo da locomoção é essencial para estabelecer métodos de tratamento de desordens do aparato locomotor. Exame da marcha representa o primeiro passo para o diagnóstico na avaliação de um paciente com claudicação. O objetivo deste estudo foi estabelecer o padrão da impressão termográfica da pegada de cães com claudicação comparada a cães saudáveis. Verificou-se também se há diferenças na temperatura da pegada de membro claudicante versus membro normal de cão normal e membro que suporta o peso de cão claudicante versus membro normal de cão normal. Foram utilizados 14 cães com claudicação unilateral do membro pélvico devido à ruptura do ligamento cruzado cranial (membro claudicante versus membro que suporta o peso) e 14 cães saudáveis (membro direito versus membro esquerdo). As imagens termográficas das pegadas foram feitas após o cão permanecer em posição ortostática sobre tapete de espuma durante 30 segundos. Foram analisados padrões térmicos de imagens e médias de temperatura. Análise da marcha foi realizada em plataforma de força na mesma posição para medir o pico de força vertical (PFV). Calculou-se média, desvio padrão, porcentagem da média e índice de assimetria para esta variável da marcha. Não houve diferença significativa na temperatura da pegada entre os grupos. Análises do padrão térmico das imagens mostraram sucesso de 80 % na diferenciação do membro claudicante contra o membro que suporta o peso (grupo anormal) e de 100% na identificação do mesmo padrão térmico entre o membro direito em relação ao esquerdo (grupo normal). A média do PFV mostrou 10,0% de diferença nos cães normais e 72,4% nos cães anormais. Análise de índice de assimetria mostrou 5% no grupo normal e 36,2% no grupo anormal. Conclui-se que a impressão termográfica da pegada pode ser utilizada com sucesso como ferramenta complementar para a detecção de claudicação dos membros pélvicos em cães.
|
114 |
Lanthanide based dendrimers for photodynamic therapy and biological optical imaging / Complexes de lanthanides formés avec de ligands dendrimères pour thérapie photodynamique et imagerie biologique optiqueNazarenko, Iuliia 17 December 2015 (has links)
La thérapie photodynamique (PDT) est une méthode de lutte contre le cancer basée sur l’utilisation de la lumière et d’un composé sensible à la lumière, appelé photosensibilisateur (PS). Le PS absorbe la lumière et, en présence d’oxygène, engendre la production des dérivés réactifs de l'oxygène (DRO), lesquels sont toxiques et provoquent la régression de la tumeur. La limitation principale des PSs utilisés dans les tests cliniques est leur faible sélectivité envers les tissus cancéreux. Le but principal de ce projet est de créer des agents multifonctionnels combinant sur une même molécule l’activité PDT, la vectorisation et l’imagerie optique proche infrarouge. Dans cette région du spectre optique, les cellules possèdent une faible autofluorescence, et la lumière proche infrarouge pénètre plus profondément dans les tissus biologiques que la lumière visible. Nous proposons ici de modifier une structure dendrimérique de type poly(amidoamine) de génération 3, en tant que plateforme polyvalente. En effet, ce dernier possède trente-deux groupes terminaux qui peuvent être facilement substitués par des PSs. De plus, cette macromolécule peut complexer dans ses cavités jusqu’à 8 cations lanthanides émettant dans le proche infrarouge. Quatre nouveaux ligands dendrimère ont été synthétisés avec différents PSs tels que des dérivés de naphtalimide, d’anthraquinone et de tétraphénylporphyrine. De plus, le naphtalimide a été couplé avec des groupes dérivés de l’acide folique pour assurer la vectorisation envers les tissus cancéreux. Les complexes de lanthanide émettant dans le proche infrarouge ont été préparés pour chaque dendrimère. La caractérisation des performances des différents complexes a été réalisée. La production de DRO et la présence de complexes d’Yb(III) a été démontrée dans les cellules HL60. Les dendrimères modifiés par les groupes anthraquinone et tétraphénylporphyrine en tant que PS, ont montré, dans les cellules vivantes, une émission proche infrarouge lorsqu’ils sont sous la forme de complexe d’Yb(III). Les résultats obtenus montrent que les complexes de lanthanides formés avec des ligands dendrimères peuvent servir comme des agents de PDT et des rapporteurs luminescents proche infrarouge in cellulo. / PDT is a cancer treatment that uses the combination of a nontoxic photoactivated molecule (photosensitizer), an appropriate source of light excitation and molecular oxygen to generate reactive oxygen species (ROS) leading to the decrease of size or to the destruction of tumors. However, the PDT efficiency of currently used drugs is limited by the selectivity for the cancer tissue. The main goal of this work is to develop a multifunctional agent which combines a PDT activity, a tumor targeting and near-infrared (NIR) optical imaging. The use of reporters that absorb at low energy is justified by low tissue autofluorescence and high tissue penetration depth in the NIR spectrum window. For this purpose, we have chosen the generation-3 poly(amidoamine) dendrimers as a versatile platform. Such macromolecules can incorporate eight NIR emitting lanthanide ions inside their branches forming species with thirty-two end groups at the periphery that can be substituted by suitable photosensitizers. Four new dendrimer ligands were synthesized with different photosensitizers, such as derivatives of naphthalimide, anthraquinone, and porphyrin. In addition the naphthalimide photosensitizer was functionalized with a targeting molecule, based on folic acid, to induce selectivity of the molecule towards cancer tissues. The corresponding NIR emitting lanthanide complexes were prepared for each dendrimer. Four Yb(III)-dendrimer complexes were characterized for their photophysical and ROS production properties. All complexes demonstrated a ROS production. The dendrimer functionalized with anthraquinone and tetraphenylporphyrin photosensitizers show strong NIR emission in living cells. These new multifunctional Yb(III)-dendrimer complexes have been designed to broaden the current scope of PDT agents and of NIR optical imaging agents.
|
115 |
Single Image Super Resolution with Infrared Imagery and Multi-Step Reinforcement LearningVassilo, Kyle January 2020 (has links)
No description available.
|
116 |
SCHLIEREN IMAGING AND INFRARED HEAT TRANSFER MEASUREMENTS ON A FLARED CONE AND CONE-CYLINDER-FLARE IN MACH-6 QUIET FLOWZachary Allen McDaniel (18431658) 26 April 2024 (has links)
<p dir="ltr">Pressure transducer, infrared heat transfer, and schlieren imaging data for a flared cone and cone-cylinder-flare in Mach 6 quiet flow are presented. Flared cone pressure transducer results show second-mode RMS values comparable to that found in prior experimental work. Second-mode frequency is found to linearly increase with increasing freestream unit Reynolds number, and frequency varies little between sensors for a given freestream unit Reynolds number. Turbulent intermittency begins to increase at a freestream unit Reynolds number 2x10<sup>6</sup>/m greater than the unit Reynolds number corresponding to peak second-mode RMS. peak RMS. High-speed schlieren imaging on the downstream section of the flared cone shows the second-mode disturbance following trends in power which correlate with PCB RMS. Infrared heat transfer results contain the azimuthal heating streak pattern observed for the flared cone in prior research, but the hot-cold-hot streak pattern is not seen due to limited model length. Streak heating occurs downstream of second-mode peak RMS over the freestream unit Reynolds number range of 6.4x10<sup>6</sup>/m to 10.4x10<sup>6</sup>/m. The heat transfer of streaks is found to vary significantly from streak to streak, while mean streak heating variation with freestream unit Reynolds number is small.</p><p dir="ltr">PCB results of the cone-cylinder-flare show intermittent turbulence at a freestream unit Reynolds number of 16.0x10<sup>6</sup>/m. Examination of shear-layer and second-mode instabilities show significant increases in RMS moving downstream along the flare and with increasing freestream unit Reynolds number. High-speed schlieren imaging of the shear-layer reattachment region on the flare show the presence of the shear-layer and second-mode instabilities when the model is configured with a sharp nose tip. The instabilities are not present with a blunt 5 mm radius nose tip. Heat transfer is observed to increase along the downstream portion of the flare. The sharp nose tip configuration has higher heat transfer rates than the 5 mm radius nose tip configuration.</p>
|
117 |
Relier les attributs de matériaux et les paramètres de procédés de fabrication à un test de contrôle qualité, une application du concept du quality by design.Yekpe, Kétsia January 2014 (has links)
Résumé : À partir de 2002, grâce à l’introduction du concept de la Qualité par la Conception (en anglais Quality by Design : QbD) par l’agence américaine des produits alimentaires et médicamenteux, l’industrie pharmaceutique a intensifié les efforts et les investissements pour permettre une libération en temps réel des lots commerciaux. Le QbD propose que la qualité soit construite dès la conception initiale du médicament plutôt que d'être évaluée à la fin de sa fabrication. Ainsi, avec l’initiative QbD, les tests de contrôle de la qualité des médicaments, réalisés après la fabrication des comprimés, peuvent être éliminés si les paramètres qui les influencent sont contrôlés.
En effet, ces tests de contrôle qualité dits traditionnels requièrent en général plusieurs heures pour leurs préparations et leurs réalisations. Tel est le cas du test de dissolution. Ce test est très consommateur de ressources matérielles et humaines. La réalisation de stratégies de contrôle pour les tests de dissolution basée sur une approche QbD pourrait être bénéfique pour l'industrie pharmaceutique.
À travers ce travail, nous avons pu :
• proposer différentes stratégies novatrices de contrôle du test de dissolution de comprimés pharmaceutiques sur la base des principes du QbD,
• apporter un nouvel éclairage sur la compréhension des phénomènes impliqués dans la dissolution de comprimés pharmaceutiques.
Les résultats de ce projet de recherche ont permis 1) la mise en évidence des paramètres critiques influençant le test de dissolution, 2) l’élaboration et l’évaluation de modèles statistiques pour les combinaisons de variation de paramètres selon un plan d’expérience préalablement conçu, 3) la corrélation du test de dissolution à des paramètres critiques de procédés de fabrication et d’attributs de matériaux grâce aux technologies d’analyse de procédés. // Abstract : With the introduction in 2002 of the concept of Quality by Design (QbD) by the Food and Drug Administration, the pharmaceutical industry intensified efforts and investments to reach real time release of commercial batches, reducing time between manufacturing and availability to the patient. QbD proposes that quality should be built in the initial design of a product rather than being assessed at the end of the tablet manufacturing. Thus, with the QbD
initiative, quality control tests of tablets like dissolution testing performed after manufacturing could be removed if the parameters impacting them are controlled.
Indeed, quality control tests such as traditional dissolution tests generally require several hours for their preparation and their realizations. Dissolution tests are time consuming, require large amounts of material and human resources. The elimination of these tests through a QbD approach could be beneficial for the pharmaceutical industry.
Thanks to this work, it was possible to :
propose different innovative strategies to control the dissolution test of pharmaceutical
tablets based on the principles of Quality by Design,
have a better understanding of this quality control test.
The main results relies on 1) the identification of critical parameters influencing the
dissolution test, 2) the development and evaluation of statistical models for the combination of variation of parameters according to an experimental design, 3) the correlation of dissolution test to critical manufacturing process parameters and attributes of materials through process analysis technology.
|
118 |
Systeme d'imagerie hybride par codage de pupille / Hybrid imaging system with wavefront codingDiaz, Frédéric 06 May 2011 (has links)
De nouveaux concepts d’imagerie permettent aux systèmes optiques d’être plus compacts et plus performants. Parmi ces nouvelles techniques, les systèmes d’imagerie hybrides par codage de pupille allient un système optique comprenant un masque de phase et un traitement numérique. La fonction de phase implantée sur le masque rend l’image insensible à un défaut du système optique, qui peut être une aberration ou de la défocalisation. Cet avantage est obtenu au prix d’une déformation connue de l’image qui est ensuite corrigée par un traitement numérique.L’étude des propriétés de ces systèmes a été effectuée en cherchant à augmenter la profondeur de champ d’un système d’imagerie. Un gain sur ce paramètre permet déjà d’envisager le relâchement de contraintes de conception optique telles que la courbure de champ, la défocalisation thermique, le chromatisme… Dans ces techniques d’imagerie, la prise en compte du bruit du capteur constitue l’un des paramètres critiques pour le choix et l’utilisation de méthodes de traitement d’image.Les travaux menés durant cette thèse ont permis de proposer une approche originale de conception conjointe de la fonction de phase du masque et de l’algorithme de restauration d’image. Celle-ci est basée sur un critère de rapport signal à bruit de l’image finale. Contrairement aux approches connues, ce critère montre qu’il n’est pas nécessaire d’obtenir une stricte invariance de la fonction de transfert du système optique. Les paramètres des fonctions de phase optimisés grâce à ce critère sont sensiblement différents de ceux usuellement proposés et conduisent à une amélioration significative de la qualité de l’image.Cette approche de conception optique a été validée expérimentalement sur une caméra thermique non refroidie. Un masque de phase binaire qui a été mis en œuvre en association avec un traitement numérique temps réel implémenté sur une carte GPU a permis d’augmenter la profondeur de champ de cette caméra d’un facteur 3. Compte-tenu du niveau de bruit important introduit par l’utilisation d’un capteur bolométrique, la bonne qualité des images obtenues après traitement démontre l’intérêt de l’approche de conception conjointe appliquée à l’imagerie hybride par codage de pupille. / New imaging techniques allow better and smaller systems. Among these new techniques, hybrid imaging systems with wavefront coding includes an optical system with a phase mask and a processing step. The phase function of the mask makes the system insensitive to a fault of the optical system, such as an aberration or a defocus. The price of this advantage is a deformation of the image acquired by a sensor, which is then processed. The study of the properties of these hybrid imaging systems has been completed by increasing the depth of field of an imaging system, which allows to relax some design constraints such as field curvature, thermal defocus, chromaticism… In these imaging techniques, the consideration the noise of the sensor is one the critical parameters when choosing the image processing method.The work performed during this thesis allowed to proposed an original approach for the cross-conception of the phase function of the mask and the processing step. This approach is based on a signal-to-noise criterion. Unlike known approaches, this criterion shows that a strict insensitivity of the modulation transfer function of the optics is not required. The parameters of the phase functions optimized thanks to this criterion are noticeably different from those usually proposed and lead to a significant increase of the image quality.This cross-conception approach has been validated experimentally on an uncooled thermal camera. A binary phase mask associated with a real-time processing implemented on a GPU allowed to increase the depth of field of this camera by a factor 3. Considering the important level of noise introduced by the use of a bolometric sensor, the good quality of the processed image shows the interest of the cross-conception for hybrid imaging system with wavefront coding.
|
119 |
3D scanning of transparent objects / Numérisation 3D d'objets transparentsEren, Gönen 22 October 2010 (has links)
Beaucoup de tâches pratiques dans l'industrie, tels que l'inspection automatique ou la vision robotique, nécessitent souvent de numérisation de formes en trois dimensions (3D) avec des techniques non-contact. Toutefois, les objets transparents, tels que ceux en verre, posent encore des difficultés pour les techniques classiques de numérisation. La reconstruction de la géométrie de surface pour les objets transparents est compliquée par le fait que la lumière est transmise à travers, réfracté et dans certains cas, réfléchie par la surface. Les approches actuelles ne peut traiter que les sous-classes d'objets. Les algorithmes sont encore très spécifiques et ne sont généralement pas applicables. En outre, de nombreuses techniques exigent un effort considérable d'acquisition et de calibration. Cette thèse propose une nouvelle méthode de détermination de la forme de la surface des objets transparents. La méthode est basée sur le chauffage locale de la surface et sur l'imagerie thermique. Tout d'abord, la surface de l'objet est chauffé avec une source laser. Une image thermique est acquis, et les coordonnées en pixels du point d'échauffement sont calculés. Ensuite, les coordonnées 3D de la surface sont déterminées en utilisant triangulation et l'étalonnage initial du système. Le processus est répété en déplaçant l'objet transparent pour reprendre sa forme de surface complète. Cette méthode est appelée "Scanning From Heating". Considérant le faisceau laser comme une source de chauffage point et la surface de l'objet localement plane à la zone d'impact, la méthode est utilisée pour obtenir les normales de la surface de l'objet, en plus des coordonnées 3D. Un prototype base sur cette méthode a été développé pendant la thèse. / Many practical tasks in industry, such as automatic inspection or robot vision, often require scanning of three-dimensional shapes with non-contact techniques. However, transparent objects, such as those made of glass, still pose difficulties for classical scanning techniques. The reconstruction of surface geometry for transparent objects is complicated by the fact that light is transmitted through, refracted and in some cases reflected by the surface. Current approaches can only deal relatively well with sub-classes of objects. The algorithms are still very specific and not generally applicable. Furthermore, many techniques require considerable acquisition effort and careful calibration. This thesis proposes a new method of determining the surface shape of transparent objects. The method is based on local surface heating and thermal imaging. First, the surface of the object is heated with a laser source. A thermal image is acquired, and pixel coordinates of the heated point are calculated. Then, the 3D coordinates of the surface are computed using triangulation and the initial calibration of the system. The process is repeated by moving the transparent object to recover its surface shape. This method is called Scanning From Heating. Considering the laser beam as a point heating source and the surface of the object locally flat at the impact zone, the Scanning From Heating method is extended to obtain the surface normals of the object, in addition to the 3D world coordinates. A scanner prototype based on Scanning From Heating method has been developed during the thesis.
|
120 |
Development and Testing of a Second Generation Hand-held Optical ImagerGonzalez, Jean 22 March 2012 (has links)
Hand-held optical imagers are developed towards clinical breast cancer imaging. Herein, a Gen-2 hand-held optical imager has been developed with unique features: (i) image curved breast tissues with ~86% surface contact, and (ii) perform reflectance and transillumination imaging using the novel forked probe heads. Extensive phantom studies were performed using 1% Liposyn solution (background, ~ 300 ml and 1000 ml volumes) and 0.45 cc India Ink (absorption) targets, under different target:background contrast ratios and target depths. Two-dimensional surface images detected target(s) up to 2.5 cm deep via reflectance imaging, and up to 5 cm deep via transillumination imaging. Preliminary studies on gel-based breast phantoms (~700 ml) detected targets via reflectance and transillumination imaging. Preliminary in-vivo reflectance studies on normal and cancerous breast tissues also detected targets, although with artifacts. In future, the portable Gen-2 imager has potential for clinical breast imaging via reflectance and transillumination approach after extensive in-vivo studies.
|
Page generated in 0.3185 seconds