La rectification est un procédé couramment utilisé dans l’industrie pour la finition des surfaces. L’optimisation du procédé consiste à trouver un compromis entre la qualité des pièces, minimiser les temps d'usinage et augmenter l’efficacité énergétique grâce au choix judicieux des paramètres de rectification. Par ailleurs le taux de production des pièces rectifiées est souvent limité par des contraintes sur la topographie de la surface et des problèmes liés à l’apparition de brûlures de rectification ou de micro-fissures à la surface des pièces. Ces défauts d’aspect engendrent généralement, lorsqu’ils concernent une surface fonctionnelle, une réduction de la durée de vie du composant ainsi rectifié.’effet des conditions de rectification et des propriétés des matériaux sur la nature des contraintes résiduelles a été analysé par modélisation numérique. Le modèle élément finis permet la prédiction non seulement des contraintes résiduelles, mais aussi des phases en présence et des déformations associées. L'objectif de cette étude est de construire un modèle numérique fiable en se basant sur la méthode des éléments finis pour analyser les contraintes résiduelles induites par la rectification et d'explorer, par conséquent, les mécanismes en termes de conditions de rectification. La variation des contraintes résiduelles et des déformations aux points d'intégration a été analysée. Les effets du coefficient de frottement (µ), du nombre de Peclet (Pe), de la conductance de paroi (H) et du flux de chaleur (Q) sur la microstructure et l’état de contraintes résiduelles ont été analysés. Enfin, sur la base des nouveaux résultats de ce travail de recherche, une méthodologie plus complète est proposée pour la suite. / Grinding is a commonly used finishing process to produce components of desired shape, size and dimensional accuracy. The ultimate goal is to have the maximum workpiece quality, minimum machining time and high economic efficiency by making a selective adaptation of the possible process strategy and chosen parameter selection. The focus of this study arose from a limitation that challenges the grinding industry. The production rate of the ground parts is generally constrained by surface topography and subsurface damage appearing as residual tensile stress, localized burns, and phase transformation induced micro and macro-cracking. This motivates the need for a reliable numerical modelling to simulate the grinding process. The numerical model sought should be able to predict not only the required grinding residual stresses but also the deformation history. The objective of this thesis is to build up a reliable finite element model for grinding-induced residual stress analysis and thus to explore thoroughly the mechanisms in terms of grinding conditions. The variations of the residual stresses and strains at integration points have been examined, and the effects of the friction coefficient (µ), Peclet number (Pe), non dimensional heat transfer coefficient (H) and different magnitudes of input heat flux (Q) on both the microstructure and the residual stress state are analyzed. Finally, based on the new findings in this research, a more comprehensive methodology is suggested for further study.

Together with an internationally praised Constitution, South Africa can pride itself in having one of the best sets of environmental laws in the world, and since EIAs have become mandatory for projects that negatively affect the environment, our track record with respect to environmental protection has substantially improved from the Apartheid days. But as the integrated and proactive approach to environmental management has seemingly been established, a new provision was introduced: Section 24G of NEMA (“Rectification of unlawful commencement of activity”), allowing for retrospective authorisation of unlawful activities (i.e. activities identified in terms of S24 of NEMA). Six years after the introduction of the S24G rectification provision in NEMA, its ability to bring about increased levels of compliance is being questioned. The findings of this research show that in Gauteng S24G has had an ambivalent effect on compliance: while on one hand it has artificially increased compliance by rendering illegal activities legal, on the other, it has seriously undermined the overall compliance and enforcement effort by opening the door to abuse and effectively providing an escape route for potential criminals. It is argued that the schizophrenic character of S24G is at the heart of this dilemma. This research presents theoretical and practical perspectives on non-compliance and analyses data collected on S24G applications in Gauteng in order to determine the effect of S24G on compliance. It also identifies key factors influencing effectiveness of S24G, and derives from the above key performance areas to improve effectiveness of S24G. / Thesis (M. Environmental Management)--North-West University, Potchefstroom Campus, 2012

Mémoire numérisé par la Division de la gestion de documents et des archives de l'Université de Montréal.

Metal-Insulator-Metal (MIM), Metal-Insulator-Insulator-Metal (MIIM), and Metal-Insulator-Insulator-Insulator-Metal (MIIIM) quantum tunneling diodes have been designed, fabricated, and characterized. The key interest of this work was to develop tunneling diodes capable of operating and detecting THz radiation up to 30THz, which is well beyond the operation ranges of other semiconductor-based diodes. Al2O3, HfO2 and TiO2 metal oxides were employed for studying the behavior of metal-insulator-metal (MIM) and metal-insulator-insulator-metal (MIIM) quantum tunneling diodes. Specifically, ultra-thin films of these oxides with varied thicknesses were deposited by atomic layer deposition (ALD) as the tunneling junction material that is sandwiched between platinum (Pt) and titanium (Ti) electrodes, with dissimilar work functions of 5.3 eV and 4.1 eV, respectively. Due to the unique and well-controlled tunneling characteristic of the ALD ultra-thin films, reproducible MIM and MIIM diode devices have been developed. The DC characteristics of MIM and MIIM tunneling junctions with different junction areas and materials were investigated in this work. The effects of the different compositions and thicknesses of the tunneling layer on the diodes were studied systematically. Through the introduction of stacked dual tunneling layers, it is demonstrated that the MIIM and MIIIM diodes exhibited a high degree of asymmetry (large ratio between forward and reverse currents) and a strong nonlinearity in their I-V characteristics. The characterization was performed on diodes with micro and nano-scale junction areas. The MIM diodes reported herein exhibited lower junction resistances than those reported by prior works. Moreover, a study was conducted to numerically extract the average barrier heights by fitting the analytical model of the tunneling current to the measured I-V responses, which were evaluated with respect to the thickness of the constituent tunneling layer. RF characterization was performed on the MIM diodes up to 65GHz, and its junction impedance was extracted. A rigorous procedure was followed to extract the diode equivalent circuit model to obtain the intrinsic lumped element model parameters of the MIM diodes.

La capture en relief d'une scène réelle peut être réalisée grâce à un couple de caméras vidéo (banc stéréoscopique). La capture de ces images vidéo stéréoscopiques et leur restitution sur des systèmes de projection en relief sont à l'interface entre les domaines de la réalité virtuelle, de la vision par ordinateur, et du cinéma en relief. Placé au sein de cette très vaste thématique, ce travail concerne la projection en relief, sur des systèmes de Réalité Virtuelle, d'images issues d'une capture par un banc stéréoscopique fixe. De très nombreuses contraintes (limitations des configurations de capture et des conditions de restitution notamment) ont restreint l'utilisation de cette technologie. Dans ce mémoire de thèse, nous détaillons les améliorations que nous avons apportées à certaines étapes de la chaîne de transmission stéréoscopique, afin de maîtriser la restitution de vidéos stéréoscopiques. Pour atteindre cet objectif, nous avons réalisé une modélisation mathématique détaillée des caméras, et des différentes configurations de capture et de restitution que nous utilisons. Disposer d'images stéréoscopiques les moins déformées possibles était un point de départ indispensable à la suite de notre travail. Dans ce but, nous avons développé un algorithme de rectification d'images vidéo stéréoscopiques. Afin d'assurer une rectification temps réel, nous avons implémenté cet algorithme sur processeur de carte graphique (GPU ou Graphics Processing Unit), en mettant en place une technique à base de table de référence. La distance interoculaire de l'utilisateur est un paramètre important pour assurer une bonne restitution des images sur les systèmes de Réalité Virtuelle. Pourtant par commodité, la valeur moyenne de cet écart est souvent prise comme référence, alors que d'importantes différences existent d'un utilisateur à l'autre. Afin d'améliorer la restitution en fixant plus précisément ce paramètre critique, nous avons développé une méthode de calibration de la distance interoculaire de l'utilisateur. Enfin, alors que les spectateurs des salles de cinéma en relief sont assis dans une zone bien définie devant l'écran, le déplacement des utilisateurs devant le système de projection d'images stéréoscopiques est une caractéristique des systèmes de Réalité Virtuelle. Pour palier aux problèmes que l'on rencontre lors de la projection d'images issues d'un banc stéréoscopique fixe pour un utilisateur en mouvement, nous proposons une méthode pour maitriser la restitution de la profondeur perçue par cet utilisateur, en nous basant sur une segmentation en profondeur de la scène.

<p>The objective of this thesis is to investigate if it is possible to use stereo vision to find and track the players and the ball during a football game.</p><p>The thesis shows that it is possible to detect all players that isn’t too occluded by another player. Situations when a player is occluded by another player is solved by tracking the players from frame to frame.</p><p>The ball is also detected in most frames by looking for ball-like features. As with the players the ball is tracked from frame to frame so that when the ball is occluded, the positions is estimated by the tracker.</p>

No description available.

GPS/INS (Global Positioning System / Inertial Navigation System) brings possibility of relaxing the demand for aerial triangulation in stereo model construction and rectification. In this thesis a differential rectification algorithm is programmed with Matlab software for aerial frame camera images. This program is tested using exterior orientation parameters obtained by GPS/INS and images are ortho-rectified. Ground Control Points (GCP) are measured in the orthorectified images and compared with other rectification methods according to RMSE and mean error. Besides, direct georeferencing accuracy is investigated by using GPS/INS data. Therefore, stereo models and ortho-images are constructed by using exterior orientation parameters obtained with both aerial triangulation and GPS/INS integration. GCPs are measured in both stereo models and ortho-images, compared with respect to their RMSE and mean error. In order to determine Digital Elevation Model (DEM) effect in ortho-rectification, different DEM data are used and the results are compared.

The purpose of this work was to construct and analyze a THz time domain spectroscopy (THz-TDS) system by using a nanojoule energy per pulse ultrafast laser (non-amplified ultrafast laser or oscillator) source and a non-linear optical generation method for THz generation. First a THz-TDS system, which uses photoconductive antenna (PCA) method for THz generation, was built to understand the working principles of these types of systems. This THz-TDS system which used PCA for generation and a 2mm thick &lt / 110&gt / ZnTe crystal for detection had a bandwidth up to 1 THz with a 1000:1 signal to noise ratio (S/N). Using this system, various materials were investigated to study the usefulness of the obtained bandwidth. Absorption coefficient and refractive indices of the sample materials were calculated. Results showed that the bandwidth of the system was not sufficient to obtain fingerprint properties of these materials. In order to improve the system, optical rectification method was used for THz generation. A different THz-TDS system was built with a 1mm thick &lt / 110&gt / ZnTe crystal used for the method of non-linear generation of THz radiation. Theoretical calculations of radiated intensity and electric field were done to analyze the expected bandwidth of the system. Results showed that the generation and the detection crystal thicknesses affect the obtained bandwidth of the system in that the bandwidth limiting factor is the crystal thickness and not the ultrafast laser pulse duration. Especially for detection, measurements obtained with both a 1mm thick and 2mm thick &lt / 110&gt / ZnTe crystal showed that there was not much difference in bandwidth as was predicted by theory. Also in order to increase the signal to noise ratio, the optics used in the system were optimized. It was found that by using same focal lengths for focusing and collimating optics around the generation crystal and by using a short focal length parabolic mirror, S/N could be improved. After these improvements this THz-TDS system which uses optical rectification for THz generation and electro-optic method for THz detection had a larger bandwidth up to 3 THz but with a lower 100:1 signal to noise ratio.

This study presents an integrated framework for the automatic generation of the photorealistic 3-d building models from satellite and ground-level imagery. First, the 2-d building patches and the corresponding footprints are extracted from a high resolution imagery using an adaptive fuzzy-genetic algorithm approach. Next, the photorealistic facade textures are automatically extracted from the single ground-level building images using a developed approach, which includes facade image extraction, rectification, and occlusion removal. Finally, the textured 3-d building models are generated automatically by mapping the corresponding textures onto the facades of the models. The developed 2-d building extraction and delineation approach was implemented on a selected urban area of the Batikent district of Ankara, Turkey. The building regions were extracted with an approximate detection rate of 93%. Moreover, the overall delineation accuracy was computed to be 3.9 meters. The developed concept for facade image extraction was tested on two distinct datasets. The facade image extraction accuracies were computed to be 82% and 81% for the Batikent and eTrims datasets, respectively. As to rectification results, 60% and 80% of the facade images provided errors under ten pixels for the Batikent and eTrims datasets, respectively. In the evaluation of occlusion removal, the average scores were computed to be 2.58 and 2.28 for the Batikent and eTrims datasets, respectively. The scores are ranked between 1 (Excellent) to 6 (Unusable). The modeling of the total 110 single buildings with the photorealistic textures took about 50 minutes of processor running time and yielded a satisfactory level of accuracy.