<p>The formation of star clusters occurs in a complex environment and involve a large number of physical processes. One of the most important processes to consider is radiative feedback. The radiation released by forming stars heats the surrounding gas and suppresses the fragmentation of low mass objects. Ionizing radiation can also drive large scale outflows and disperse the surrounding gas. Owing to all this complexity, the use of numerical simulations to study cluster formation in molecular clouds has become commonplace. In order to study the effects of radiative feedback on cluster formation over larger spatial scales than previous studies, we present hydrodynamical simulations using the AMR code FLASH which make use of cluster particles. Unlike previous studies, these particles represent an entire star cluster rather than individual stars. We present a subgrid model for representing the radiative output of a star cluster which involves randomly sampling an IMF over time to populate the cluster. We show that our model is capable of reproducing the properties of observed clusters. The model was then incorporated into FLASH to examine the effects of radiative feedback on cluster formation in full hydrodynamical simulations. We find that the inclusion of radiative transfer can drive large scale outflows and decreases the overall star formation efficiency by a factor of 2. The inclusion of radiative feedback also increases the degree of subclustering. The use of cluster particles in hydrodynamical simulations represents a promising method for future studies of cluster formation and the large scale effects of radiative feedback.</p> / Master of Science (MSc)

Yes / Recovering high-quality 3D point clouds from monocular endoscopic images is a challenging task. This paper proposes a novel deep learning-based computational framework for 3D point cloud reconstruction from single monocular endoscopic images. An unsupervised mono-depth learning network is used to generate depth information from monocular images. Given a single mono endoscopic image, the network is capable of depicting a depth map. The depth map is then used to recover a dense 3D point cloud. A generative Endo-AE network based on an auto-encoder is trained to repair defects of the dense point cloud by generating the best representation from the incomplete data. The performance of the proposed framework is evaluated against state-of-the-art learning-based methods. The results are also compared with non-learning based stereo 3D reconstruction algorithms. Our proposed methods outperform both the state-of-the-art learning-based and non-learning based methods for 3D point cloud reconstruction. The Endo-AE model for point cloud completion can generate high-quality, dense 3D endoscopic point clouds from incomplete point clouds with holes. Our framework is able to recover complete 3D point clouds with the missing rate of information up to 60%. Five large medical in-vivo databases of 3D point clouds of real endoscopic scenes have been generated and two synthetic 3D medical datasets are created. We have made these datasets publicly available for researchers free of charge. The proposed computational framework can produce high-quality and dense 3D point clouds from single mono-endoscopy images for augmented reality, virtual reality and other computer-mediated medical applications.

Energy conservation is a major concern in todays data centers, which are the 21st century data processing factories, and where large and complex software systems such as distributed data management stores run and serve billions of users. The two main drivers of this major concern are the pollution impact data centers have on the environment due to their waste heat, and the expensive cost data centers incur due to their enormous energy demand. Among the many subsystems of data centers, the storage system is one of the main sources of energy consumption. Among the many types of storage systems, key/value stores happen to be the widely used in the data centers. In this work, I investigate energy saving techniques that enable a consistent hash based key/value store save energy during low activity times, and whenever there is an opportunity to reuse the waste heat of data centers.

Energy conservation is a major concern in todays data centers, which are the 21st century data processing factories, and where large and complex software systems such as distributed data management stores run and serve billions of users. The two main drivers of this major concern are the pollution impact data centers have on the environment due to their waste heat, and the expensive cost data centers incur due to their enormous energy demand. Among the many subsystems of data centers, the storage system is one of the main sources of energy consumption. Among the many types of storage systems, key/value stores happen to be the widely used in the data centers. In this work, I investigate energy saving techniques that enable a consistent hash based key/value store save energy during low activity times, and whenever there is an opportunity to reuse the waste heat of data centers.

This thesis is about visualizing a kind of data that is trivial to process by computers but difficult to imagine by humans because nature does not allow for intuition with this type of information: high-dimensional data. Such data often result from representing observations of objects under various aspects or with different properties. In many applications, a typical, laborious task is to find related objects or to group those that are similar to each other. One classic solution for this task is to imagine the data as vectors in a Euclidean space with object variables as dimensions. Utilizing Euclidean distance as a measure of similarity, objects with similar properties and values accumulate to groups, so-called clusters, that are exposed by cluster analysis on the high-dimensional point cloud. Because similar vectors can be thought of as objects that are alike in terms of their attributes, the point cloud\'s structure and individual cluster properties, like their size or compactness, summarize data categories and their relative importance. The contribution of this thesis is a novel analysis approach for visual exploration of high-dimensional point clouds without suffering from structural occlusion. The work is based on implementing two key concepts: The first idea is to discard those geometric properties that cannot be preserved and, thus, lead to the typical artifacts. Topological concepts are used instead to shift away the focus from a point-centered view on the data to a more structure-centered perspective. The advantage is that topology-driven clustering information can be extracted in the data\'s original domain and be preserved without loss in low dimensions. The second idea is to split the analysis into a topology-based global overview and a subsequent geometric local refinement. The occlusion-free overview enables the analyst to identify features and to link them to other visualizations that permit analysis of those properties not captured by the topological abstraction, e.g. cluster shape or value distributions in particular dimensions or subspaces. The advantage of separating structure from data point analysis is that restricting local analysis only to data subsets significantly reduces artifacts and the visual complexity of standard techniques. That is, the additional topological layer enables the analyst to identify structure that was hidden before and to focus on particular features by suppressing irrelevant points during local feature analysis. This thesis addresses the topology-based visual analysis of high-dimensional point clouds for both the time-invariant and the time-varying case. Time-invariant means that the points do not change in their number or positions. That is, the analyst explores the clustering of a fixed and constant set of points. The extension to the time-varying case implies the analysis of a varying clustering, where clusters appear as new, merge or split, or vanish. Especially for high-dimensional data, both tracking---which means to relate features over time---but also visualizing changing structure are difficult problems to solve.

Approved for public release, distribution is unlimited / The Monterey Bay sea breeze varies because of the influence of features such as inversions, clouds, synopticscale flow, and topography. The sea breeze is important because it impacts fire weather, air pollution, agriculture, and aviation operations, among other things. Analyses are conducted using a multi-quadric based program, which incorporates aircraft data, surface observations, and profiler data, to investigate the Monterey Bay sea breeze during 01-31 August 2003. Factors including inversions, cloud cover, amount of heating, distribution of heating, synoptic-scale flow, and topography are studied to determine their influence on the sea breeze. Six days are selected that best illustrate the factors that influence the structure of the Monterey Bay sea breeze. Results show that offs hore flow weakened the strength of the sea breeze and decreased the depth. A cooling trend in surface temperatures at the end of August also weakened the strength of the sea breezes and decreased the depth. Clouds are present during this period, which influenced the amount of heating, and consequently, the sea breeze response. The presence of a marine layer weakened the thermal gradient that in turn, weakened the sea breeze circulation. / Lieutenant Junior Grade, United States Naval Reserve

Approved for public release, distribution is unlimited / The objective of this thesis is to obtain a better understanding of cloud-top entrainment through an in-depth analysis of entrainment-zone structure. In situ aircraft measurements taken during the Atlantic Stratocumulus Transition Experiment (ASTEX) were used for this purpose. Using data collected from multiple cloud-top penetrations, the presence of an interfacial layer in-between the top of the cloud mixed-layer and the base of the free atmosphere is identified and consequently defined as the entrainment zone. The depth of the entrainment zone is on the order of tens of meters, where turbulence and sometimes cloud droplets are detectable. Inhomogeneous mixing was found to occur within the entrainment zone. Parcels of inversion-layer air and boundary-layer air are identified within the entrainment zone. Analyses suggest that turbulence intensity and cloud amount in the entrainment zone vary depending on the distribution of entrainment mixing fraction. Furthermore, continuous mixing in the entrainment zone appears to dissipate the upper-cloud layer. However, continuous dissipation of the upper-cloud layer has not been observed. Further study is needed to determine the interaction between cloud-top entrainment and the full integration of boundary-layer dynamics. / Captain, United States Air Force

De nombreux scanners laser permettent d'obtenir la surface 3D a partir d'un objet. Néanmoins, la surface reconstruite est souvent lisse, ce qui est du au débruitage interne du scanner et aux décalages entre les scans. Cette these utilise des scans haute precision et choisit de ne pas perdre ni alterer les echantillons initiaux au cours du traitement afin de les visualiser. C'est en effet la seule façon de decouvrir les imperfections (trous, decalages de scans). De plus, comme les donnees haute precision capturent meme le plus leger detail, tout debruitage ou sous-echantillonnage peut amener a perdre ces details.La these s'attache a prouver que l'on peut trianguler le nuage de point initial en ne perdant presque aucun echantillon. Le probleme de la visualisation exacte sur des donnees de plus de 35 millions de points et de 300 scans differents est ainsi resolu. Deux problemes majeurs sont traites: le premier est l'orientation du nuage de point brut complet et la creation d'un maillage. Le second est la correction des petits decalages entre les scans qui peuvent creer un tres fort aliasing et compromettre la visualisation de la surface. Le second developpement de la these est une decomposition des nuages de points en hautes/basses frequences. Ainsi, des methodes classiques pour l'analyse d'image, l'arbre des ensembles de niveau et la representation MSER, sont etendues aux maillages, ce qui donne une methode intrinseque de segmentation de maillages. Une analyse mathematiques d'operateurs differentiels discrets, proposes dans la litterature et operant sur des nuages de points est realisee. En considerant les developpements asymptotiques de ces operateurs sur une surface reguliere, ces operateurs peuvent etre classifies. Cette analyse amene au developpement d'un operateur discret consistant avec Ie mouvement par courbure moyenne (l'equation de la chaleur intrinseque) definissant ainsi un espace-echelle numerique simple et remarquablement robuste. Cet espace-echelle permet de resoudre de maniere unifiee tous les problemes mentionnes auparavant (orientation et triangulation du nuage de points, fusion de scans, segmentation de maillages) qui sont ordinairement traites avec des techniques distinctes. / Many laser devices acquire directly 3D objects and reconstruct their surface. Nevertheless, the final reconstructed surface is usually smoothed out as a result of the scanner internal de-noising process and the offsets between different scans. This thesis, working on results from high precision scans, adopts the somewhat extreme conservative position, not to loose or alter any raw sample throughout the whole processing pipeline, and to attempt to visualize them. Indeed, it is the only way to discover all surface imperfections (holes, offsets). Furthermore, since high precision data can capture the slightest surface variation, any smoothing and any sub-sampling can incur in the loss of textural detail.The thesis attempts to prove that one can triangulate the raw point cloud with almost no sample loss. It solves the exact visualization problem on large data sets of up to 35 million points made of 300 different scan sweeps and more. Two major problems are addressed. The first one is the orientation of the complete raw point set, an the building of a high precision mesh. The second one is the correction of the tiny scan misalignments which can cause strong high frequency aliasing and hamper completely a direct visualization.The second development of the thesis is a general low-high frequency decomposition algorithm for any point cloud. Thus classic image analysis tools, the level set tree and the MSER representations, are extended to meshes, yielding an intrinsic mesh segmentation method.The underlying mathematical development focuses on an analysis of a half dozen discrete differential operators acting on raw point clouds which have been proposed in the literature. By considering the asymptotic behavior of these operators on a smooth surface, a classification by their underlying curvature operators is obtained.This analysis leads to the development of a discrete operator consistent with the mean curvature motion (the intrinsic heat equation) defining a remarkably simple and robust numerical scale space. By this scale space all of the above mentioned problems (point set orientation, raw point set triangulation, scan merging, segmentation), usually addressed by separated techniques, are solved in a unified framework.

This thesis is devoted to the experimental quantitative characterization of the shape and orientation distribution of ice particles in clouds. The characterization is based on measured and modeled elevation dependencies of the polarimetric parameters differential reflectivity and correlation coefficient. The polarimetric data is obtained using a newly developed 35-GHz cloud radar MIRA-35 with hybrid polarimetric configuration and scanning capabilities. The full procedure chain of the technical implementation and the realization of the setup of the hybrid-mode cloud radar for the shape determination are presented. This includes the description of phase adjustments in the transmitting paths, the introduction of the general data processing scheme, correction of the data for the differences of amplifications and electrical path lengths in the transmitting and receiving channels, the rotation of the polarization basis by 45°, the correction of antenna effects on polarimetric measurements, the determination of spectral polarimetric variables, and the formulation of a scheme to increase the signal-to-noise ratio. Modeling of the polarimetric variables is based on existing backscattering models assuming the spheroidal representation of cloud scatterers. The parameters retrieved from the model are polarizability ratio and degree of orientation, which can be assigned to certain particle orientations and shapes. In the thesis the first quantitative estimations of ice particle shape at the top of liquid-topped clouds are presented. Analyzed ice particles were formed in the presence of supercooled water and in the temperature range from -20 °C to -3 °C. The estimation is based on polarizability ratios of ice particles measured by the MIRA-35 with hybrid polarimetric configuration, manufactured by METEK GmbH. For the study, 22 cases observed during the ACCEPT (Analysis of the Composition of Clouds with Extended Polarization Techniques) field campaign were used. Polarizability ratios retrieved for cloud layers with cloud-top temperatures of about -5, -8, -15, and -20 °C were 1.6, 0.9, 0.6, and 0.9, respectively. Such values correspond to prolate, quasi-isotropic, oblate, and quasi-isotropic particles, respectively. Data from a free-fall chamber were used for the comparison. A good agreement of detected shapes with well-known shape{temperature dependencies observed in laboratories was found.

Neste trabalho estudamos a geometria do campo magnético e algumas propriedades da poeira interestelar em duas regiões do Meio Interestelar da Galáxia: a Nuvem Escura de Musca e uma parte da IRAS Vela Shell, na direção de HD62542. Empregamos a técnica da polarimetria de imagem CCD. Na primeira parte desta tese mostramos nosso aporte no aprimoramento do hardware e software do grupo de polarimetria do IAGUSP. Basicamente, um código de controle da gaveta polarimétrica do IAG foi desenvolvido e explicamos em detalhe um pacote (PCCDPACK) de redução semi-automática e análise de imagens polarimétricas para objetos pontuais criado especificamente para este fim. Na segunda parte desta tese aplicamos a técnica de polarimetria de imagem CCD para um estudo observacional em duas regiões selecionadas: a nuvem escura de musca (NEM) e a região IRAS Vela Shell (IVS) / nebulosa de Gum (NG) na direção de HD62542. Nosso interesse foi explorar a técnica polarimétrica desenvolvida para a redução e análise de campos estelares densos. O produto deste estudo foi a construção de dois catálogos polarimétricos, um para cada região, de ao redor de 2500 objetos para a NEM e quase 900 objetos para a região IVS/NG. Mostramos também como tais medidas podem ser utilizadas para mapear o campo magnético em regiões densas do meio interestelar, especificamente em nuvens escuras e frentes de ionização/shocks onde a presença de grãos de poeira é prevista. Informamos sobre a intensidade do campo magnético e das razões de energia cinética turbulenta a magnética é obtida do estudo das dispersões do ângulo de polarização dos vetores de polarização em cada uma das regiões. Paralelamente, um estudo de extinção utilizando a técnica de contagem automática de estrelas foi feito em cada uma das regiões e sua correlações com os dados polarimétricos são exploradas. Combinando as medidas polarimétricas com as de extinção, a eficiência da polarização é ) investigada assim como suas implicâncias na determinação do tipo de grãos presentes ao serem comparados com aqueles próprios do meio interestelar difuso. A análise da polarimetria na NEM mostra um limite inferior de polarização de aprox. 2% ao longo da estrutura filamentar da nuvem, com picos de polarização de 6-7% na região central. A geometria do campo magnético, como é inferida do mapa de polarização, é quase perpendicular ao filamento, sugerindo um colapso ao longo das linhas do campo. No entanto, uma análise em pequena escala angular mostra variações no ângulo de polarização de aprox. 30% ao redor do valor médio de 110º. O campo magnético estimado ao longo da nuvem se encontra dentro de uma faixa de 0.05 mgauss a o.30 mgauss. Um limite inferior para a massa da nuvem de 139 M SOL é encontrado e um grande numero de condensações foram detectadas na estrutura filamentar com uma escala típica de tamanho \'L\' aprox. 0.26pc. As correlações entre polarização e extinção sugerem que a poeira na NEM possui diferentes propriedades que as do meio interestelar difuso e que uma privilegiada visão geométrica do campo magnético pode estar presente ao longo da nuvem. O estudo da região IVS/NG permitiu estabelecer uma geometria do campo magnético que parece acompanhar, em algumas regiões, a borda da parede de emissão Halfa. No entanto, em outras regiões, parece evidente uma tendência perpendicular parede. O campo magnético estimado ao longo da frente de ionização está na faixa de 0.02 mgauss a 0.11 mgauss e a pressão magnética parece dominar a pressão turbulenta do gás em algumas regiões. Uma evidência de estruturas tipo clumpy foi detectada com limites inferiores típicos de (1-4) masas solares e uma escala de tamanho de L\' aprox. 0.47pc. A eficiência da polarização muda ao longo da frente de ionização e parece claro que em algumas regiões as propriedades da poeira diferem daquelas ) observadas no meio interestelar difuso. Um ótimo alinhamento é observado em algumas regiões o que sugere uma ótima visão geométrica do campo magnético com respeito à frente de ionização vista edge on. / We have used CCD imaging polarimetry to study the geometry of the magnetic field and some properties of the interstellar dust in two regions of the Interstellar Medium of the Galaxy: the Musca Dark Cloud and a section of the IRAS Vela Shell towards HD62542. We initially describe the instrument employed with the Musca cloud, the IAG polarimeter, for which we developed control software. A very similar instrument was used at CTIO for the Vela Shell. We have also developed an IRAF software package to reduce and analyze polarimetric images in crowded fields (PCCDPACK). We next apply the technique to study two selected regions: the Musca Dark Cloud (MDC) and a section of IRAS Vela Shell (IVS). The end product is the construction of two catalogues, of approximately 2500 objects for MDC and 900 objects for IVS. For MDC, the analyses of the polarimetric data show a polarization lower limit of ~2% along the filamentary structure of the cloud, with peaks of 6-7% in the central regions. The geometry of the magnetic field is approximately perpendicular to filament, suggesting collapse along the field lines. However, when looked in detail, we detect variations in the polarization angle of ~30deg respect to the mean value of 110deg. The estimated strength of the magnetic field is in a range of 0.05-0.30 mgauss. The star count technique yields a lower limit of the total mass of the cloud of 139 solar masses. A large number of condensations are found, with a typical length scale L ~ 0.26 pc. The observed correlation between polarization and extinction suggests that the dust in MDC have different properties with respect to the dust in the diffuse Interstellar Medium. It also appears that the magnetic field along the cloud is viewed favorably so as to produce the observed polarization. The study in the section of the IVS towards HD 62542 allows us to detect a magnetic field parallel to the ionization front in sections of the cloud. However, in others regions, a perpendicular geometry is also evident. The estimated magnetic field is in the range of 0.02-0.11 mgauss and the magnetic pressure may dominate the turbulent pressure of gas in some regions. Evidence of clumpy structure is found with typical masses (1-4) solar masses and a length scale L ~ 0.47 pc. The polarization efficiency changes along the ionization front. It is clear that the properties of the dust are different of the interstellar medium in some regions. An optimum alignment of the polarization vectors is seen in some regions and it may reflect a favorable viewing geometry of the magnetic field with respect of the ionization front seeing edge on.