The problem of reconstruction of digital images from their degraded measurements has always been a problem of central importance in numerous applications of imaging sciences. In real life, acquired imaging data is typically contaminated by various types of degradation phenomena which are usually related to the imperfections of image acquisition devices and/or environmental effects. Accordingly, given the degraded measurements of an image of interest, the fundamental goal of image reconstruction is to recover its close approximation, thereby "reversing" the effect of image degradation. Moreover, the massive production and proliferation of digital data across different fields of applied sciences creates the need for methods of image restoration which would be both accurate and computationally efficient. Developing such methods, however, has never been a trivial task, as improving the accuracy of image reconstruction is generally achieved at the expense of an elevated computational burden. Accordingly, the main goal of this thesis has been to develop an analytical framework which allows one to tackle a wide scope of image reconstruction problems in a computationally efficient manner. To this end, we generalize the concept of variable splitting, as a tool for simplifying complex reconstruction problems through their replacement by a sequence of simpler and therefore easily solvable ones. Moreover, we consider two different types of variable splitting and demonstrate their connection to a number of existing approaches which are currently used to solve various inverse problems. In particular, we refer to the first type of variable splitting as Bregman Type Splitting (BTS) and demonstrate its applicability to the solution of complex reconstruction problems with composite, cross-domain constraints. As specific applications of practical importance, we consider the problem of reconstruction of diffusion MRI signals from sub-critically sampled, incomplete data as well as the problem of blind deconvolution of medical ultrasound images. Further, we refer to the second type of variable splitting as Fuzzy Clustering Splitting (FCS) and show its application to the problem of image denoising. Specifically, we demonstrate how this splitting technique allows us to generalize the concept of neighbourhood operation as well as to derive a unifying approach to denoising of imaging data under a variety of different noise scenarios.

<p>The focus of this work is to use geologic and geophysical methods to better understand the faults and fracture systems at Puna, in southeastern Hawaii, and southern Montserrat, in the Lesser Antilles. The particular interest is understanding and locating the deep fracture networks that are necessary for fluid circulation in hydrothermal systems. The dissertation first presents a study in which identification of large scale faulting places Montserrat into a tectonic context. Then follow studies of Puna and Montserrat that focus on faults and fractures of the deep hydrothermal systems.</p><p>The first chapter consists of the results of the SEA-CALIPSO experiment seismic reflection data, recorded on a 48 channel streamer with the active source as a 2600 in3 airgun. This chapter discusses volcaniclastic debris fans off the east coast of Montserrat and faults off the west coast. The work places Montserrat in a transtensional environment (influenced by oblique subduction) as well as in a complex local stress regime. One conclusion is that the stress regime is inconsistent with the larger arc due to the influence of local magmatism and stress.</p><p>The second chapter is a seismic study of the Puna hydrothermal system (PHS) along the Kilauea Lower East Rift Zone. The PHS occurs at a left step in the rift, where a fracture network has been formed between fault segments. It is a productive geothermal field, extracting steam and reinjecting cooled, condensed fluids. A network of eight borehole seismometers recorded >6000 earthquakes. Most of the earthquakes are very small (< M.2), and shallow (1-3 km depth), likely the result of hydrothermal fluid reinjection. Deeper earthquakes occur along the rift as well as along the south-dipping fault plane that originates from the rift zone.</p><p>Seismic methods applied to the PHS data set, after the initial recording, picking, and locating earthquakes, include a tomographic inversion of the P-wave first arrival data. This model indicates a high seismic velocity under the field that is thought to be an intrusion and the heat source of the hydrothermal system. A shear wave splitting study suggested the PHS fracture system is largely oriented rift-parallel with some orthogonal fractures. Shear wave splitting data also were used in a tomographic inversion for fracture density. The fracture density is high in the PHS, which indicates high permeability and potential for extensive fluid circulation. This has been confirmed by high fluid flow and energy generation. The high fracture density is consistent with the interpretation of a transfer zone between the rift segments where a fracture mesh would be expected. In Puna the transfer zone is a relay ramp.</p><p>The results from the PHS are used as an example to examine the proposed hydrothermal system at St. George's Hill, Montserrat. In southern Montserrat, hot springs and fumaroles suggest a deep hydrothermal system heated by local magmatism. A magnetotelluric study obtained resistivity data that suggest focused alteration under southeastern Montserrat that is likely to be along fault segments. Several faults intersect under SGH, making it the probable center of the hydrothermal system. At Puna, and also Krafla, Iceland, where faults interact is an area of increased permeability, acting as a model to be applied to southern Montserrat. The conclusion is that in both Puna and Montserrat large faults interact to produce local areas of stress transfer that lead to fracturing and permeable networks; these networks allow for high-temperature hydrothermal circulation.</p> / Dissertation

The thesis studies the relationship between inter-fiber forces present within a cellulose fiber web under varying external conditions. It particularly concentrates on the degree of fiber to fiber bonding and fiber entanglement as a function of moisture content. Finite element analysis of the fiber bonding is used in conjunction with the experimental results to analyze and explain the inter-fiber behaviors taking place within a sheet.

ABSTRACT In Taiwan, to spin-off the OEM from OBM (Owned-Brand Manufacture) business has been a hot topic of disputed discussion for a long time since Stan Shih announced his strategic policy in ACER reorganization in 2000. Starting from then, even the slightly internal reorganization, the officially spin off WISTRON from ACER under the Taiwan¡¦s Corporate M&A (Merger and Acquisition) Law which was launched in 2002, the ACER and WISTRON have got success respectively later on, and several similar reorganization case of OEM and OBM business splitting in Taiwan ICT electronic industry and et cetera, all have triggered a lot of disputed discussion. And the arguing has never been ceased. The advantages and disadvantages of splitting OEM from OBM business have been researched and discussed by many experts. Either to-do or not-to-do the spinoff has its own devotees. No consensus has been made at all. Regardless the determination of spinoff has put the enterprise in a dilemma, but this paper focuses on the key success factors in management of splitting OEM from OBM business in Taiwan ICT electronic firms from the perspective of human resources and corporate relationship management. Because the issues and concerning arising from splitting OEM from OBM are not wholly covered under the scope of enterprise¡¦s Re-engineering at scale¡¦s down-scoping. The difficulties of splitting OEM from OBM business are more than the difficulties of down-scoping Re-engineering which is just mainly emphasizing at Refocusing. It needs not only to consider increasing competitive advantage on both side of OEM and OBM but also be faced with how to carefully handle the response of the interested parties, including stake holders, employees, customers, venders, government and community organizations and media and more. Via the methodology of Case Study, this paper intends to understand those Taiwan's information and communication technology (ICT) electronic firms who chose splitting OEM from OBM business while facing of global environmental change, how they re-organized in order to refocus business and built a lean supply chain. In the mean while, from the view point of human resources manage and corporate relationship management, explores what the key success factors shall be handled in the implementation of the reorganization splitting OEM from OBM business. Key words: OEM/ODM, OBM/Owned-Brand, Spin-off/splitting, KSF/Key success factor

The modeling of nuclear reactors involves the solution of a multi-physics problem with widely varying time and length scales. This translates mathematically to solving a system of coupled, non-linear, and stiff partial differential equations (PDEs). Multi-physics applications possess the added complexity that most of the solution fields participate in various physics components, potentially yielding spatial and/or temporal coupling errors. This dissertation deals with the verification aspects associated with such a multi-physics code, i.e., the substantiation that the mathematical description of the multi-physics equations are solved correctly (both in time and space). Conventional paradigms used in reactor analysis problems employed to couple various physics components are often non-iterative and can be inconsistent in their treatment of the non-linear terms. This leads to the usage of smaller time steps to maintain stability and accuracy requirements, thereby increasing the overall computational time for simulation. The inconsistencies of these weakly coupled solution methods can be overcome using tighter coupling strategies and yield a better approximation to the coupled non-linear operator, by resolving the dominant spatial and temporal scales involved in the multi-physics simulation. A multi-physics framework, KARMA (K(c)ode for Analysis of Reactor and other Multi-physics Applications), is presented. KARMA uses tight coupling strategies for various physical models based on a Matrix-free Nonlinear-Krylov (MFNK) framework in order to attain high-order spatio-temporal accuracy for all solution fields in amenable wall clock times, for various test problems. The framework also utilizes traditional loosely coupled methods as lower-order solvers, which serve as efficient preconditioners for the tightly coupled solution. Since the software platform employs both lower and higher-order coupling strategies, it can easily be used to test and evaluate different coupling strategies and numerical methods and to compare their efficiency for problems of interest. Multi-physics code verification efforts pertaining to reactor applications are described and associated numerical results obtained using the developed multi-physics framework are provided. The versatility of numerical methods used here for coupled problems and feasibility of general non-linear solvers with appropriate physics-based preconditioners in the KARMA framework offer significantly efficient techniques to solve multi-physics problems in reactor analysis.

This study investigates the accuracy and efficiency of several flux splitting methods for the compressible, two-dimensional Euler equations. Steger-Warming flux vector splitting method, Van Leer flux vector splitting method, The Advection Upstream Splitting Method (AUSM), Artificially Upstream Flux Vector Splitting Scheme (AUFS) and Roe&rsquo / s flux difference splitting schemes were implemented using the first- and second-order reconstruction methods. Limiter functions were embedded to the second-order reconstruction methods. The flux splitting methods are applied to subsonic, transonic and supersonic flows over NACA0012 airfoil, as well as subsonic, transonic and supersonic flows in a channel. The comparison of the obtained results with each other and the ones in the literature is presented. The advantages and disadvantages of each scheme among others are identified.

This thesis aimed to examine the process of projective identification in the couple relationship by focusing on two areas of investigation: Firstly, how projective identification operates &ldquo / within&rdquo / an individual regarding the associations among early parenting experiences, personality and couple relationship / secondly how projective identification operates between two partners in the couple relationship regarding partners&rsquo / similarities and complementarities have been focused. Initially, several multiple regressions were run to examine the relationships among the parenting, personality and couple relationship. Afterwards, several intra-class partial pairwise correlatons were conducted to reveal similarities and complementarities of the partners regarding their early maladaptive parenting experiences, personality constructs and couple relationship variables. Paulson Daily Living Inventory, Separation-Individuation Inventory and Splitting Scale were translated into Turkish and psychometric studies were conducted. For the main study, 356 participants (178 male and 178 female), who are cohabiting partners to each other, were achieved through snowball sampling method. Early maladaptive parenting experiences of the participants were examined by Young Parenting Inventory. Personality constructs contained &ldquo / separation individuation process&rdquo / and &ldquo / splitting defense&rdquo / as the theoretical correlates of projective identification, and &ldquo / early maladaptive schemas&rdquo / of the individuals. Couple relationship variables were &ldquo / relationship satisfaction&rdquo / , &ldquo / emotional dependency&rdquo / and &ldquo / jealousy&rdquo / . Findings of the study revealed that projective identification has bidirectional relatedness with the personality and couple relationship of the individual. Multiple regression analyses showed theoretically consistent associations among an individual&rsquo / s parenting experiences, personality and couple relationship. Intra-class partial pairwise correlations showed similarities and complementarities between two partners. Findings were discussed in the light of relevant literature.

This dissertation consists of two essays on the execution cost of financial markets. In the first essay, we study impacts of new block trading rules on two kinds of large trades, block trades (BTs) and splitting order trades (STs). We find some results with policy implications. First, targets traded in the block trading market are illiquid. The proportion of BTs (STs) is the decreasing (increasing) function of stock liquidity. Second, large orders are mainly executed by STs except for illiquid stocks, but investors prefer to trade in block trading market at times when trade size is large, probability of informed trading is lower and price volatility is mild. Third, under the new system, the conditional execution costs of BTs (STs) decline (do not decline) and the percentage of BTs (STs) increases (decreases).Fourth, BTs are uninformed with motivations of tax minimization, general trades and ownership transfer trades, while STs are information based. BTs for tax minimization (general trades) incur the lowest (highest) execution costs. Uninformed BTs incur higher execution costs than informed STs, reflecting ¡§premiums of trading with the specific counterparty¡¨ of BTs. Finally, simulation analyses confirm that the block trading market functions well especially for illiquid stocks. In the second essay, we attempt to provide evidence regarding the welfare effect of pre-trade transparency affected by investor and order types. In order to understand the effect of transparency on welfare, we need to explore investors¡¦ behavior adjustments (aggressiveness and order size adjustments) reacted to transparency. We find both individual and institutional investors are more willing to supply liquidity after transparency enhancement. Individual investors behave more aggressively and submit larger order as they supply and demand liquidity, while institutional investors are relatively conservative and submit smaller order in an open environment. We measure welfare of investors in terms of implementation shortfall, which is weighted average of price impacts and opportunity costs. Our main result is, on average, institutional and individual investors who demand immediacy benefit from pre-trade transparency, especially for institutional investors, while traders who supply liquidity are worse off except institutional investors. Intraday analysis further notices individual investors providing liquidity near the end of day lose most from transparency enhancement, while institutional and individual investors demanding liquidity win most in close interval.

The design, fabrication, and characterizations of the spin-polarized AlxGa1-xN/GaN HEMT structure have been achieved for spintronic application. By band calculation within linear combination of atomic orbitals and two-band k·p methods, the theoretical spin-splitting energy and minimum-spin-splitting surface of wurtzite structure have been investigated as a function of the Fermi wavevector with various strain-relaxations. Base on these results, the design of host material of the nonballistic spin-FET has also been proposed. By optimizing the Al composition and n2DEG, the Fermi surface of two-dimensional electron gas is supposed to reach the minimum-spin-splitting surface to produce resonant spin-lifetime. Because the high quality AlxGa1-xN/GaN HEMT structure is necessary for realizing the spin-FET, the influence of the growth conditions on the polarity and structure quality of the GaN epilayer have been studied on the sample grown by plasma-assisted molecular beam epitaxy. Ga-polar AlGaN/GaN heterostructures on c-Al2O3 has been realized by growing over the Al-rich AlN nucleation layer. And the reduction of interface roughness and threading dislocation scatterings of the electrons in two-dimensional electron gas has also been achieved by growing GaN epilayer under slightly Ga-rich condition. Furthermore, the effect of different types of threading dislocation on the electron mobility of the AlxGa1-xN/GaN HEMT structure has been investigated as well. At low temperature, the electron mobility of two-dimensional electron gas in AlGaN/GaN heterostructures is majorly scattered by the edge type dislocation rather than the screw type. The designs of proposed host material for spin-FETs have been realized through growing high quality spin-polarized AlxGa1-xN/GaN HEMT structures with various Al composition (x= 0.191 ¡V 0.397) grown on c-Al2O3 by metalorganic vapor phase epitaxy. The high mobility (10682 cm2/Vs at 0.4 K), flat interface (surface roughness < 0.5 nm), and high quality HEMT provide a good environment to study the spin-splitting energy. To investigate the spin-splitting energy as functions of the Fermi wavevector, the Shubnikov-de Haas measurements were performed. A large spin-splitting energy (10.76 meV) has been fabricated in Al0.390Ga0.61N/GaN HEMT structure with kf = 8.14 ¡Ñ 108 m-1 for the host material of the Datta-Das spin-FET. And for the first time, the minimum-spin-splitting surface has been experimentally generated in Al0.390Ga0.61N/GaN HEMT structure with kf = 8.33 ¡Ñ 108 m-1 for the host material of the nonballistic spin-FET.

In this dissertation, we propose using and analytically evaluate the predictive multicast polling scheme and the tree splitting algorithm for medium access control in interference dominating wireless access networks with random traffic and finite nodes. In an interference dominating wireless network, a receiver could simultaneously receive multiple packets from a variety of transmitters, as long as the signal-to-interference-plus-noise ratio exceeds a predetermined threshold. We concentrate on the case of in which the maximum queue size in a node is finite. We use discrete-time Markov chains, reward processes and regenerative processes to derive the throughput, the packet blocking probability, the average packet delay, and the average system size. We show that the system performance of the predictive multicast polling scheme can be significantly improved with a few additional buffers in the queues. Our study also shows that exact performance of the splitting algorithm depends on the total number of nodes in the networks. We verify our numerical results by rigorous mathematical proof and computer simulations.