No description available.

This dissertation describes the design and implementation of a system that has been designed to extract 3D information from pairs of 2D images. System input consists of two images taken by an ordinary digital camera. System output is a full 3D model extracted from 2D images. There are no assumptions about the positions of the cameras during the time when the images are being taken, but the scene must not undergo any modifications. The process of extracting 3D information from 2D images consists of three basic steps. First, point matching is performed. The main contribution of this step is the introduction of an approach to matching image segments in the context of an approximation space. The second step copes with the problem of estimating external camera parameters. The proposed solution to this problem uses 3D geometry rather than the fundamental matrix widely used in 2D to 3D conversion. In the proposed approach (DirectGS), the distances between reprojected rays for all image points are minimised. The contribution of the approach considered in this step is a definition of an optimal search space for solving the 2D to 3D conversion problem and introduction of an efficient algorithm that minimises reprojection error. In the third step, the problem of dense matching is considered. The contribution of this step is the introduction of a proposed approach to dense matching of 3D object structures that utilises the presence of points on lines in 3D space. The theory and experiments developed for this dissertation demonstrate the usefulness of the proposed system in the process of digitizing 3D information. The main advantage of the proposed approach is its low cost, simplicity in use for an untrained user and the high precision of reconstructed objects.

The production of high quality silicon requires the use of ultraclean coal containing less than 1.5% ash. The magnetite used to clean the coal in a dense medium process is a contaminant that seriously impacts the quality of the final silicon product. As such, research has been conducted to evaluate the potential to substitute the magnetite with fine silica–based alternative material generated during the silicon production process. Dense medium cyclone tests were performed based on a statistically designed program to determine the optimum conditions that maximize organic efficiency and minimize probable error and low–density bypass. The results revealed that a clean coal product with less than 1.5% ash can be produced using a medium formed from the silicon production waste with an organic efficiency value of around 99% and a probable error value below 0.02. There was no measurable bypass of high density particles into the product stream or low–density particles into the reject stream.

Dans un moteur à propergol solide, l'écoulement dépend fortement des gouttes d'alumine en suspension, dont la fraction massique est élevée. La distribution en taille des gouttes, qui s'élargit avec la coalescence, joue un rôle clef. Or résoudre des écoulements diphasiques polydisperses instationnaires avec une bonne précision sur la taille est un défi à la fois sur le plan de la modélisation et du calcul scientifique: (1) de très petites gouttes, par exemple résultant de la combustion de nanoparticules d'aluminium, subissent mouvement brownien et coalescence, (2) de petites gouttes ont leur vitesse conditionnée par leur taille de sorte qu'elles coalescent lorsqu'elles ont des tailles différentes, (3) des gouttes plus grosses peuvent se croiser par effet d'inertie et (4) toutes les gouttes interagissent de manière fortement couplée avec la phase porteuse. En complément des approches lagrangiennes, des modèles eulériens ont été développés pour décrire la phase dispersée à un coût raisonnable, et ils permettent un couplage aisé avec la phase porteuse ainsi que la parallélisation massive des codes: les approches eulériennes sont bien adaptées aux calculs industriels. Le modèle Multi-Fluide permet la description détaillée de la polydispersion, des coreélations taille/vitesse et de la coalescence, en résolvant séparément des "fluides" de gouttes triées par taille, appelés sections. Un ensemble de modèles est évalué dans cette thèse et une stratégie numérique est développée pour effectuer des calculs industriels de moteurs à propergol solide. (1) La physique des nanoparticules est évalué et incluse dans un modèle de coalescence complet. Des méthodes de moments d'ordre élevé sont ensuite développées: (2) une méthode à deux moments en taille est étendue à la coalescence pour traiter la physique de la polydispersion et les développements numériques connexes permettent d'effectuer des calculs applicatifs dans le code industriel CEDRE; (3) une méthode basée sur les moments en vitesse du deuxième ordre, un schéma de transport à l'ordre deux sur maillages structurés ainsi qu'un modèle de coalescence sont développés. Des validations académiques de la stratégie pour gouttes d'inertie modérée sont effectuées sur des écoulements complexes puis avec de la coalescence; (4) une stratégie d'intégration en temps est développée et mise en œuvre dans CEDRE pour traiter efficacement le couplage fort, dans des cas instationnaires et polydisperses incluant de très petites particules. L'ensemble des développements est soigneusement validé: soit par des formules analytiques ad hoc pour la coalescence et pour le couplage fort d'une onde acoustique; soit par des comparaisons numériques croisées avec une DPS pour la coalescence et avec des simulations lagrangiennes de cas applicatifs, coalescents et fortement couplés; soit par des résultats expérimentaux disponibles sur une configuration académique de coalescence et sur un tir de moteur à échelle réduite. La stratégie complète permet des calculs applicatifs à un coût raisonnable. En particulier, un cal- cul de moteur avec des nanoparticules permet d'évaluer la faisabilité de l'approche et d'orienter les efforts de recherche sur les propergols chargés de nanoparticules.

Dans un moteur à propergol solide, l'écoulement dépend fortement des gouttes d'alumine en suspension, dont la fraction massique est élevée. La distribution en taille des gouttes, qui s'élargit avec la coalescence, joue un rôle clef. Or résoudre des écoulements diphasiques polydisperses instationnaires avec une bonne précision sur la taille est un défi à la fois sur le plan de la modélisation et du calcul scientifique: (1) de très petites gouttes, par exemple résultant de la combustion de nanoparticules d'aluminium, subissent mouvement brownien et coalescence, (2) de petites gouttes ont leur vitesse conditionnée par leur taille de sorte qu'elles coalescent lorsqu'elles ont des tailles différentes, (3) des gouttes plus grosses peuvent se croiser par effet d'inertie et (4) toutes les gouttes interagissent de manière fortement couplée avec la phase porteuse. En complément des approches lagrangiennes, des modèles eulériens ont été développés pour décrire la phase dispersée à un coût raisonnable, et ils permettent un couplage aisé avec la phase porteuse ainsi que la parallélisation massive des codes: les approches eulériennes sont bien adaptées aux calculs industriels. Le modèle Multi-Fluide permet la description détaillée de la polydispersion, des coreélations taille/vitesse et de la coalescence, en résolvant séparément des "fluides" de gouttes triées par taille, appelés sections. Un ensemble de modèles est évalué dans cette thèse et une stratégie numérique est développée pour effectuer des calculs industriels de moteurs à propergol solide. (1) La physique des nanoparticules est évalué et incluse dans un modèle de coalescence complet. Des méthodes de moments d'ordre élevé sont ensuite développées: (2) une méthode à deux moments en taille est étendue à la coalescence pour traiter la physique de la polydispersion et les développements numériques connexes permettent d'effectuer des calculs applicatifs dans le code industriel CEDRE; (3) une méthode basée sur les moments en vitesse du deuxième ordre, un schéma de transport à l'ordre deux sur maillages structurés ainsi qu'un modèle de coalescence sont développés. Des validations académiques de la stratégie pour gouttes d'inertie modérée sont effectuées sur des écoulements complexes puis avec de la coalescence; (4) une stratégie d'intégration en temps est développée et mise en œuvre dans CEDRE pour traiter efficacement le couplage fort, dans des cas instationnaires et polydisperses incluant de très petites particules. L'ensemble des développements est soigneusement validé: soit par des formules analytiques ad hoc pour la coalescence et pour le couplage fort d'une onde acoustique; soit par des comparaisons numériques croisées avec une DPS pour la coalescence et avec des simulations lagrangiennes de cas applicatifs, coalescents et fortement couplés; soit par des résultats expérimentaux disponibles sur une configuration académique de coalescence et sur un tir de moteur à échelle réduite. La stratégie complète permet des calculs applicatifs à un coût raisonnable. En particulier, un cal- cul de moteur avec des nanoparticules permet d'évaluer la faisabilité de l'approche et d'orienter les efforts de recherche sur les propergols chargés de nanoparticules.

Many researches have shown that urban parks can enhance the quality of life, which is a key factor of building a sustainable city. Hong Kong, as one of the most densely populated financial city in the world, has severely low urban park density. Instead, a large number of small public urban open spaces (SPUOS) are scattered within the city fabrics. This paper therefore aims to study the design and functions of these SPUOS. It is also hoped that find out whether a network of connected small open spaces can compensate for a large urban park. After reviewing the history of urban parks in Hong Kong. It is believed that the multitude of SPUOS is the result of poor urban planning and imperfect urban renewal schemes during the early occupation by the British Government. These SPUOS are usually around 1000 to 1500 m2 large in size and can be abundantly found along adjacent streets. For the design of the SPUOS, benches are usually the only facility installed and the vegetation is of poor quality. The connectivity of SPUOS located within the study area in the Yau Tsim Mong District is investigated with the use of graph theory and connectivity indices. The result has shown that these SPUOS in the study area have a high degree of connectivity. They are further compared with a larger urban park located nearby. Based on the comparisons and other prior researches, a conclusion has been drawn that size of a park is not the major concern of park users but rather its facilities and design. It is believed that a network of connected SPUOS might serve the local community better due to their locations. However, they should not be able to entirely replace large urban parks since they have fewer amenities provided. The SPUOS are found to have rather different functions to the society than what urban parks are commonly expected. Their environment and economic functions are constrained mainly by the poor design and the locations. An interview with the representative of a non-government organization has been made to further understand their social functions. The interviewee believes that SPUOS now mainly act as social hubs and living rooms for senior citizens. This statement is then verified by the result of a prior survey and some other researches. This is due to the combination of aging population structure, exploding population growth, and expanding poverty gap. Lastly, a comparative study has been made on the small public urban green spaces (SPUGS) in Copenhagen. The SPUGS can attract visitors of different age group from other far neighborhoods. Despite than fact that there is cultural difference and citizens from these two cities have different lifestyles, a better design of the pocket park might be the reason of why it becomes an attractive natural meeting point in the central Copenhagen for all age-groups visitors. Based on all the investigations and findings, suggestions have been made to improve the quality of SPUOS in Hong Kong.

This dissertation describes the design and implementation of a system that has been designed to extract 3D information from pairs of 2D images. System input consists of two images taken by an ordinary digital camera. System output is a full 3D model extracted from 2D images. There are no assumptions about the positions of the cameras during the time when the images are being taken, but the scene must not undergo any modifications. The process of extracting 3D information from 2D images consists of three basic steps. First, point matching is performed. The main contribution of this step is the introduction of an approach to matching image segments in the context of an approximation space. The second step copes with the problem of estimating external camera parameters. The proposed solution to this problem uses 3D geometry rather than the fundamental matrix widely used in 2D to 3D conversion. In the proposed approach (DirectGS), the distances between reprojected rays for all image points are minimised. The contribution of the approach considered in this step is a definition of an optimal search space for solving the 2D to 3D conversion problem and introduction of an efficient algorithm that minimises reprojection error. In the third step, the problem of dense matching is considered. The contribution of this step is the introduction of a proposed approach to dense matching of 3D object structures that utilises the presence of points on lines in 3D space. The theory and experiments developed for this dissertation demonstrate the usefulness of the proposed system in the process of digitizing 3D information. The main advantage of the proposed approach is its low cost, simplicity in use for an untrained user and the high precision of reconstructed objects.

A stereo vision system is a robust method to sense the distance information in a scene. This research explores the stereo vision system from the fundamentals of stereo vision and the computer stereo vision algorithm to the final implementation of the system on a FPGA chip. In a stereo vision system, images are captured by a pair of stereo image sensors. The distance information can be derived from the disparities between the stereo image pair, based on the theory of binocular geometry. With the increasing focus on 3D vision, stereo vision is becoming a hot topic in the areas of computer games, robot vision and medical applications. Particularly, most stereo vision systems are expected to be used in real-time applications. In this thesis, several stereo correspondence algorithms that determine the disparities between stereo image pair are examined. The algorithms can be categorized into global stereo algorithms and local stereo algorithms depending on the optimization techniques. The global algorithms examined are the Dynamic Time Warp (DTW) algorithm and the DTW with quantization algorithm, while the local algorithms examined are the window based Sum of Squared Differences (SSD), Sum of Absolute Differences (SAD) and Census transform correlation algorithms. With analysis among them, the window based SAD correlation algorithm is proposed for implementation on a FPGA platform. The proposed algorithm is implemented onto an Altera DE2 board featuring an Altera Cyclone II 2C35 FPGA. The implemented module of the algorithm is simulated using ModelSim-Altera to verify the correctness of its functionality. Along with a pair of stere image sensors and a LCD monitor, a stereo vision system is built. The entire system realizes a real-time video frame rate of 16.83 frames per second with an image resolution of 640 by 480 and produces disparity maps in which the objects are clearly distinguished by their relative distance information.

For the last two decades, the possibility of using synthetic membranes for industrial gas separations has attracted considerable interest since membrane separation technologies have the advantages of energy efficiency, simplicity and low cost. However, for wider commercial utilization there is still a need to develop membranes with higher permeant fluxes and higher transport selectivities. Conductive polymers, due to their high gas transport selectivities, give rise to a new class of polymeric materials for membrane based gas separation though poor mechanical properties obstruct the applications for this purpose of use. This problem led researches to a new idea of combining the conducting polymers with insulating polymers forming mixed matrix composite membranes. In the previous studies in our group, polypyrrole was chosen as the conductive polymer, and different preparation techniques were tried and optimized for membrane application. As the insulating polymer, previously poly(bisphenol-Acarbonate) was used to support the conductive polymer filler in order to constitute a conductive composite membrane. For this study, as the polymer matrix, hexafluorobisphenol A based poly(arylene ether ketone) was targeted due to its physical properties and temperature resistance which can be important for industrial applications. First of all, permeabilities of N2, CH4, Ar, H2, CO2, and H2 were measured at varying temperatures ranging from 25&deg / C to 85&deg / C through a homogenous dense membrane of chosen polymeric material to characterize its intrinsic properties. Measurements were done using laboratory scale gas separation apparatus which makes use of a constant volume variable pressure technique. The permeability results were used for the calculations of permeation activation energies for each gas. These permeation activation energies were found to be differing slightly for each gas independently from the kinetic diameters of gases. In this study, mixed matrix membranes of conducting polymer, polypyrrole (PPy) and insulating polymer, hexafluorobisphenol A based poly(arylene ether ketone) (PAEK) were also prepared. It was observed that PAEK and PPy form a composite mixed matrix structure, which can function as permselective membrane. The effect of conducting polymer filler content was investigated with two different filler ratios. When comparing with the pure PAEK membranes, meaningful increases for both permeability and selectivity were obtained for some of the gases.

Tetrachloroethene (PCE) is a major groundwater contaminant that often persists as dense, nonaqueous phase liquids (DNAPLs) in subsurface environments. Dissolved-phase PCE plumes emanate from DNAPL source zones, which act as continuous sources of contamination for decades. Removal of DNAPL source zones is crucial to achieve lasting remedy of contaminated aquifers. This research explored the contributions of the microbial reductive dechlorination process (i.e., anaerobic bioremediation) to PCE-DNAPL source zone remediation, either in isolation or as a polishing step for the removal of residual DNAPL remaining after application of surfactant enhanced aquifer remediation (SEAR), an emerging physical-chemical source zone treatment. Specific objectives of this research were to: (1) evaluate the ability of microorganisms to dechlorinate in the presence of PCE-DNAPL and at high dissolved-phase PCE concentrations expected near/in DNAPL source zones, (2) assess the distribution and activity of key dechlorinating populations during bioenhanced PCE-DNAPL dissolution in continuous-flow column experiments, (3) determine the influence of Tween 80, a biodegradable surfactant commonly used in SEAR, on the microbial reductive dechlorination process, (4) design and optimize quantitative real-time PCR (qPCR) protocols to detect and enumerate key dechlorinating populations (e.g., Geobacter lovleyi, Sulfurospirillum multivorans), and (5) explore the effects of oxygen on Dehalococcoides viability and biomarker quantification. This research demonstrated that microbial dechlorinating activity within DNAPL source zones promotes bioenhanced dissolution although many dechlorinating isolates cannot tolerate saturated PCE concentrations. Application of newly designed qPCR protocols established a direct link between dissolution enhancement and the distribution of relevant dechlorinating populations in the vicinity of PCE-DNAPL. The limited and reversible impact of Tween 80 on key dechlorinators supported the feasibility of a treatment train approach of SEAR followed by microbial reductive dechlorination to remediate PCE-DNAPL source zones. Finally, experiments with oxygen-exposed, Dehalococcoides-containing cultures suggested limitations of using Dehalococcoides DNA and RNA biomarkers for monitoring bioremediation at field sites. These findings advance the scientific understanding of the microbial reductive dechlorination process and are relevant to environmental remediation practitioners. The advantages and current shortcomings of PCE-DNAPL source zone bioremediation, as well as recommendations for future research, are discussed.