<div>Nucleate boiling is an efficient heat transfer mechanism that enables the dissipation of high heat fluxes at low temperature differences. Heat transfer phenomena during nucleate boiling are closely linked to the two-phase flow morphology that evolves in time and based on the operating conditions. In particular, the critical heat flux, which is the upper limit for the nucleate boiling regime, can be triggered by hydrodynamic mechanisms resulting from interactions between the liquid and vapor phases. The aim of this thesis is to characterize the two-phase flow morphology evolution during nucleate boiling at high heat fluxes in two configurations: pool boiling, and confined and submerged two-phase jet impingement. The characterization is performed via non-invasive, high-speed optical based diagnostic tools. </div><div>Experimental characterization of liquid-vapor interfaces during boiling is often challenging because the rapidly evolving vapor structures are sensitive to invasive probes and multiple interfaces can occlude one another along a line of sight. In this thesis, a liquid-vapor interface reconstruction technique based on high-speed stereo imaging is developed. Images are filtered for feature enhancement and template matching is used for determining the correspondence of local features of the liquid-vapor interfaces between the two camera views. A sampling grid is overlaid on the reference image and windows centered at each sampled pixel are compared with windows centered along the epipolar line in the target image to obtain a correlation signal. To enhance the signatures of true matches, the correlation signals for each sampled pixel are averaged over a short time ensemble correlation. The three-dimensional coordinates of each matched pixel are determined via triangulation, which yields a set of points in the physical world representing the liquid-vapor interface. The developed liquid-vapor interface reconstruction technique is a high-speed, flexible and non-invasive alternative to the various existing methods for phase-distribution mapping. This technique also has the potential to be combined with other optical-based diagnostic tools, such as tomographic particle image velocimetry, to further understand the phase interactions.<br></div><div>The liquid-vapor interface reconstruction technique is used to characterize liquid-vapor interfaces above the heated surface during nucleate pool boiling, where the textured interface resulting from the boiling phenomena and flow interactions near the heated surface is particularly suited for reconstruction. Application of the reconstruction technique to pool boiling at high heat fluxes produces a unique quantitative characterization of the liquid-vapor interface morphology near heated surface. Analysis of temporal signals extracted from reconstructions indicate a clear transition in the nature of the vapor flow dynamics from a plume-like vapor flow to a release mode dominated by vapor burst events. Further investigation of the vapor burst events allows identification of a characteristic morphology of the vapor structures that form above the surface that is associated to the square shape of the heat source. Vapor flow morphology characterization during pool boiling at high heat fluxes can be used to inform vapor removal strategies that delay the occurrence of the critical heat flux during pool boiling.</div><div>As compared to pool boiling, nucleate boiling can be sustained up to significantly higher heat fluxes during two-phase jet impingement. The increases in critical heat flux are explained via hydrodynamic mechanisms that have been debated in the literature. The connection between two-phase flow morphology and the extension of nucleate boiling regime is investigated for a single subcooled jet of water that impinges on a circular heat source via high-speed visualization from two synchronized top and side views of the confinement gap. When boiling occurs under subcooled exit flow conditions and at moderate heat fluxes, the regular formation and collapse of vapor structures that bridge the heated surface and the orifice plate is observed, which causes significant oscillations in the pressure drop across. Under saturated exit flow conditions, the vapor agglomerates in the confinement gap into a bowl-like vapor structure that recurrently shrinks, due to vapor break-off at the edge of the orifice plate, and replenishes due to vapor generation. The optical visualizations from the top of the confinement gap provide a unique perspective and indicate that the liquid jet flows downwards through the vapor structure, impinges on the heated surface, and then flows underneath the vapor structure, as a fluid wall jet the keeps the heated surface wetted such that discrete bubbles continue to nucleate. At high heat fluxes, intense vapor generation causes the fluid wall jet to transition from a bubbly to a churn-like regime, and some liquid droplets are sheared off into the vapor structure. The origin of critical heat flux appears to result from a significant portion of the liquid in the wall jet being deflected off the surface, and the remaining liquid film on the surface drying out before reaching the edge of the heater.</div><div>The flow morphology characterizations presented in this dissertation further the understanding of flow and heat transfer phenomena during nucleate boiling. In the pool boiling configuration, the vapor release process was quantitatively described; during two-phase jet impingement, a possible mechanism for critical heat flux was identified. Opportunities for future work include the utilization of image processing techniques to extract quantitative measurements from two-phase jet impingement visualizations. Also, the developed liquid-vapor interface reconstruction technique can be applied to a boiling situation with a simpler liquid-vapor interface geometry, such as film boiling, to generate benchmark data for validation and development of numerical models.</div><div><br></div>

"Patient motion, which causes artifacts in reconstructed images, can be a serious problem in Single Photon Emission Computed Tomography (SPECT) imaging. If patient motion can be detected and quantified, the reconstruction algorithm can compensate for the motion. A real-time multi-threaded Visual Tracking System (VTS) using optical cameras, which will be suitable for deployment in clinical trials, is under development. The VTS tracks patients using multiple video images and image processing techniques, calculating patient motion in three-dimensional space. This research aimed to develop and implement an algorithm for feature matching and stereo location computation using multiple cameras. Feature matching is done based on the epipolar geometry constraints for a pair of images and extended to the multiple view case with an iterative algorithm. Stereo locations of the matches are then computed using sum of squared distances from the projected 3D lines in SPECT coordinates as the error metric. This information from the VTS, when coupled with motion assessment from the emission data itself, can provide a robust compensation for patient motion as part of reconstruction."

Dans le modèle cosmologique actuel, l'Univers est majoritairement composé de matière noire dont la nature est inexpliquée par le Modèle Standard de la physique des particules. L'annihilation de particules issues de nouveaux cadres théoriques, peut induire un signal de rayons gamma de très hautes énergies, observable par des expériences d'astronomie gamma. Largement dominées par la matière noire, les galaxies naines sphéroïdes sont des cibles privilégiées pour conduire cette recherche. Le réseau de télescopes H.E.S.S. discuté dans la première partie est un parfait exemple d'expérience d'imagerie atmosphérique stéréoscopique permettant de conduire la recherche de matière noire. Une nouvelle méthode de discrimination des gerbes électromagnétiques et hadroniques permettant d'améliorer la recherche de sources faibles est présentée. Elle est appliquée aux données des observations de la galaxie naine du Sagittaire et la limite supérieure sur le flux de gamma en provenance de cet objet est calculée. En orbite à bord du satellite Fermi depuis Juin 2008, le télescope à conversion de paire LAT permet de rechercher la matière noire sur l'ensemble de la voûte céleste. La sensibilité théorique à détecter un signal de matière noire est déterminée pour deux galaxies naines spéciques. Au terme de la première année d'observations, les limites supérieures sur les ux de gamma sont dérivées pour un catalogue de galaxies naines sphéroïdes. Des modèles de physique au-delà du Modèle Standard sont confrontés avec les contraintes calculées sur les sections ecaces d'annihilation en fonction de la masse des particules obtenues à partir des observations effectuées avec ces deux expériences.

<p>3D fractals can be visualized as 3D objects with complex structure and has unlimited details. This thesis will be about methods to render 3D fractals effectively and efficiently, both to explore it in real-time and to create beautiful high resolution images with high details. The methods discussed is direct volume rendering with ray-casting and cut plane rendering to explore the fractal and an approach that uses super sampling to create high resolution images. Stereoscopic rendering is discussed and how it enhance the visual perception of the fractal</p>

Multiview video is one of the emerging research areas especially among the video coding community. Transmission of multiview video over an error prone network is possible with efficient compression of these videos. But along with the studies for efficiently compressing the multiview video, new error concealment and error protection methods are also necessary to overcome the problems due to erroneous channel conditions in practical applications. In packet switching networks, packet losses may lead to block losses in a frame or the loss of an entire frame in an encoded video sequence. In recent years several algorithms are proposed to handle the loss of an entire frame efficiently. However methods for full frame losses in stereoscopic or multiview videos are limited in the literature. In this thesis a stereoscopic approach for full frame loss concealment of multiview video is proposed. In the proposed methods, the redundancy and disparity between the views and motion information between the previously decoded frames are used to estimate the lost frame. Even though multiview video can be composed of more than two views, at most three view are utilized for concealment. The performance of the proposed algorithms are tested against monoscopic methods and the conditions under which the proposed methods are superior are investigated. The proposed algorithms are applied to both stereoscopic and multiview video.

This thesis is motivated by the necessity to understand the flow structure of turbulent flows in bends encountered in many technical applications such as heat exchangers, nuclear reactors and internal combustion engines. Flows in bends are characterised by strong secondary motions in terms of counter-rotating vortices (Dean cells) set up by a centrifugal instability. Specifically the thesis deals with turbulent flows in 90° curved pipes of circular cross-section with and without an additional motion, swirling or pulsatile, superposed on the primary flow.  The aim of the present thesis is to study these complex flows in detail by using time-resolved stereoscopic particle image velocimetry to obtain the three-dimensional velocity field, with complementary hot-wire anemometry and laser Doppler velocimetry measurements. In order to analyse the vortical flow field proper orthogonal decomposition (POD) is used. The so called ``swirl-switching'' is identified and it is shown that the vortices instantaneously, ``rock'' between three states, viz. a pair of symmetric vortices or a dominant clockwise or counter-clockwise Dean cell. The most energetic mode exhibits a single cell spanning the whole cross-section and ``rolling'' (counter-)clockwise in time. However, when a honeycomb is mounted at the inlet of the bend, the Dean vortices break down and there is strong indication that the ``swirl-switching'' is hindered. When a swirling motion is superimposed on the incoming flow, the Dean vortices show a tendency to merge into a single cell with increasing swirl intensity. POD analysis show vortices which closely resemble the Dean cells, indicating that these structures co-exist with the swirling motion. In highly pulsating turbulent flow at the exit of a curved pipe, the vortical pattern is diminished or even eliminated during the acceleration phase and then re-established during the deceleration. In order to investigate the effect of pulsations and curvature on the performance of a turbocharger turbine, highly pulsating turbulent flow through a sharp bend is fed into the turbine. Time-resolved pressure and mass-flow rate measurements show that the hysteresis loop in the pressure-ratio-mass-flow plane, may differ significantly between straight and curved inlets, however the mean operating point is only slightly affected. / <p>QC 20140523</p>

Αντικείμενο της παρούσας εργασίας είναι η υψομετρική ανίχνευση θέσης κινούμενου ρομπότ. Για τον υπολογισμό της θέσης του ρομπότ χρησιμοποιήθηκε ένα στερεοσκοπικό σύστημα δύο ίδιων παράλληλων καμερών. Οι εσωτερικές παράμετροι των δύο καμερών και το μήκος της βασικής γραμμής είναι γνωστά. Θεωρείται για ευκολία ότι όλος ο όγκος του ρομπότ αναπαριστάται απο ένα σημείο P με συντεταγμένες (X,Y,Z) ώς προς σύστημα συντεταγμένων με αρχή το κέντρο προβολής της αριστερής κάμερας. Για τον εντοπισμό του ρομπότ αποκτήθηκαν δύο ακολουθίες βίντεο, μια για κάθε κάμερα, στις οποίες έγινε κατάτμηση κίνησης με την μέθοδο διαφοράς διαδοχικών καρέ. Στη συνέχεια ακολούθησε επεξεργασία εικόνας στα καρέ και των δύο ακολουθιών για την εξαγωγή των περιοχών που απεικονίζουν το αντικείμενο στο επίπεδο της εικόνας. Κατόπιν, υπολογίστηκαν οι θέσεις των κέντρων βάρους των περιοχών για τις δύο ακολουθίες , Pl(xl,yl) και Pr(xr,yr) αντίστοιχα. Τα σημεία αυτά αναπαριστούν το σημείο P στην αριστερή και στη δεξιά ακολουθία εικόνων. Ακολούθησε αντιστοίχιση σημείων μεταξύ των δύο ακολουθιών με βάση την επιπολική γεωμετρία. Τα σημεία στα οποία βρέθηκε αντιστοιχία χρησιμοποιήθηκαν για τον υπολογισμό της θέσης του σημείου P με την μέθοδο της τριγωνοποίησης. / The object of this thesis is the position detection of a moving robot. To calculate the position of the robot a stereoscopic system of two identical parallel cameras was used. The intrinsic parameters of the cameras and the length of the baseline are given. To simplify the problem it is assumed that the entire robot can be represented by a point P with (X,Y,Z) coordinates with respect to the coordinate system of the left camera center of projection. To detect the robot two video image sequences were acquired, one for each camera, to which motion segmentation was applied with the frame differencing method. After that, image proseccing was implemented to the frame sequences in order to find the region that represents the robot in the image plane. The centroids of the regions were calculated in the two sequences,Pl(xl,yl) and Pr(xr,yr) respectively. Those points represent the point P in the image planes of the left and right camera. Finally the restriction of epipole geometry was used to find matches between those points. Using those matches the coordinates of P were calculated with the triangulation method.

The purpose of this experimental study was to investigate the effect of stereoscopic three-dimensional (S3D) images on productive and receptive recall of foreign language vocabulary. S3D images are highly-realistic and differ from non-stereoscopic three-dimensional (NS3D) images in that they provide the impression of the added third dimension of depth. This within-subject study exposed the participants (N = 82) in a controlled setting to a series of carefully designed and randomly distributed NS3D and S3D images. The subjects were then given immediate productive and receptive tests of foreign language vocabulary items that were represented by NS3D and S3D images. Quantitative data consisted of the scores from the vocabulary tests. Qualitative data, gathered through background questionnaires and follow-up surveys, included a mixture of open-ended and Likert questions. The statistical analyses of the data using a series of paired t-tests showed NS3D and S3D images to be equally effective for vocabulary recall. In addition, significantly more subjects found S3D images to be engaging and/or more useful, while subjects also indicated that they perceived the main benefits of learning with S3D images to come from enhanced focus, realism, engagement, and association. At the same time, some learners reported being distracted and experiencing discomfort while viewing S3D images. Post hoc tests revealed that lower performance on S3D images was driven only by those subgroups that exhibited discomfort and / or lack of experience with S3D technology.

Made available in DSpace on 2016-06-02T19:06:22Z (GMT). No. of bitstreams: 1 6596.pdf: 13694392 bytes, checksum: e1c7e2c6a28770665631b8b56c490c57 (MD5) Previous issue date: 2014-12-16 / Aid to navigation systems can be used to help the visually impaired to perform daily tasks, such as their mobility and orientation. The objective of this project is to develop a prototype of an aid to navigation system for visually impaired people in indoors environments. The system has video sensors that detect obstacles ahead and markers containing points of interest that may be useful to the user, as doors and stairs. The methodology of this work involves selecting tools, computer vision techniques and digital image processing that may be useful to implement a prototype to validate the proposed work. As a result, a prototype was created that detects obstacles in real-time as well as points of interest. / Os sistemas de auxílio à navegação podem ser utilizados para auxiliar deficientes visuais a realizar tarefas do cotidiano, como por exemplo sua locomoção e orientação. O objetivo deste projeto é desenvolver um protótipo de um sistema de auxílio à navegação para deficientes visuais em ambientes internos. O sistema possui sensores de vídeo que detectam obstáculos à frente, e marcadores contendo pontos de interesse que possam ser de utilidade ao usuário como portas e escadas. A metodologia deste trabalho implica em selecionar ferramentas e técnicas de visão computacional e processamento digital de imagens que possam ser de utilidade à implementação de um protótipo que valide o trabalho proposto. Como resultado, foi criado um protótipo que detecta obstáculos em tempo real, assim como pontos de interesse.

3D fractals can be visualized as 3D objects with complex structure and has unlimited details. This thesis will be about methods to render 3D fractals effectively and efficiently, both to explore it in real-time and to create beautiful high resolution images with high details. The methods discussed is direct volume rendering with ray-casting and cut plane rendering to explore the fractal and an approach that uses super sampling to create high resolution images. Stereoscopic rendering is discussed and how it enhance the visual perception of the fractal