Theses and Dissertations from UMD
Permanent URI for this communityhttp://hdl.handle.net/1903/2
New submissions to the thesis/dissertation collections are added automatically as they are received from the Graduate School. Currently, the Graduate School deposits all theses and dissertations from a given semester after the official graduation date. This means that there may be up to a 4 month delay in the appearance of a give thesis/dissertation in DRUM
More information is available at Theses and Dissertations at University of Maryland Libraries.
Browse
6 results
Search Results
Item Compressed Sensing Beyond the IID and Static Domains: Theory, Algorithms and Applications(2017) Kazemipour, Abbas; Wu, Min; Babadi, Behtash; Electrical Engineering; Digital Repository at the University of Maryland; University of Maryland (College Park, Md.)Sparsity is a ubiquitous feature of many real world signals such as natural images and neural spiking activities. Conventional compressed sensing utilizes sparsity to recover low dimensional signal structures in high ambient dimensions using few measurements, where i.i.d measurements are at disposal. However real world scenarios typically exhibit non i.i.d and dynamic structures and are confined by physical constraints, preventing applicability of the theoretical guarantees of compressed sensing and limiting its applications. In this thesis we develop new theory, algorithms and applications for non i.i.d and dynamic compressed sensing by considering such constraints. In the first part of this thesis we derive new optimal sampling-complexity tradeoffs for two commonly used processes used to model dependent temporal structures: the autoregressive processes and self-exciting generalized linear models. Our theoretical results successfully recovered the temporal dependencies in neural activities, financial data and traffic data. Next, we develop a new framework for studying temporal dynamics by introducing compressible state-space models, which simultaneously utilize spatial and temporal sparsity. We develop a fast algorithm for optimal inference on such models and prove its optimal recovery guarantees. Our algorithm shows significant improvement in detecting sparse events in biological applications such as spindle detection and calcium deconvolution. Finally, we develop a sparse Poisson image reconstruction technique and the first compressive two-photon microscope which uses lines of excitation across the sample at multiple angles. We recovered diffraction-limited images from relatively few incoherently multiplexed measurements, at a rate of 1.5 billion voxels per second.Item Multi-scale problems on collective dynamics and image processing(2014) Tan, Changhui; Tadmor, Eitan; Applied Mathematics and Scientific Computation; Digital Repository at the University of Maryland; University of Maryland (College Park, Md.)Multi-scale problems appear in many contexts. In this thesis, we study two dif- ferent subjects involving multi-scale problems: (i) collective dynamics, and (ii) image processing. For collective dynamics, we concentrate on flocking models, in particular, Cucker-Smale and Motsch-Tadmor systems. These models characterize the emergent behaviors of self-organized dynamics. We study flocking systems in three different scales, from microscopic agent-based models, through mesoscopic kineitc discriptions, to macroscopic fluid systems. Global existence theories are developed for all three scales, with the proof of asymptotic flocking behaviors. In the macroscopic level, a critical threhold phenomenon is addressed to obtain global regularity. Similar idea is implemented to other fluid systems as well, like Euler-Poisson equations. In the kinetic level, a discontinuous Galerkin method is introduced to overcome the numerical difficulty due to the precence of δ -singularity. For image processing, we apply the idea of multi-scale image representation to construct uniformly bounded solutions for div U = F. Despite the fact that the equation is simple and linear, it is suprisingly true that its bounded solution can not be constructed through a linear procedure. In particular, the Holmholtz solution is not always bounded. A hierarchical construction of the bounded solution of the equation is proposed, borrowing the idea from image processing. We also present a numerical implementation to deal with the highly nonlinear construction procedure. Solid numerical result verifies that the constructed solution is indeed uniformly bounded.Item SCHLIEREN SEQUENCE ANALYSIS USING COMPUTER VISION(2013) Smith, Nathanial Timothy; Lewis, Mark J; Aerospace Engineering; Digital Repository at the University of Maryland; University of Maryland (College Park, Md.)Computer vision-based methods are proposed for extraction and measurement of flow structures of interest in schlieren video. As schlieren data has increased with faster frame rates, we are faced with thousands of images to analyze. This presents an opportunity to study global flow structures over time that may not be evident from surface measurements. A degree of automation is desirable to extract flow structures and features to give information on their behavior through the sequence. Using an interdisciplinary approach, the analysis of large schlieren data is recast as a computer vision problem. The double-cone schlieren sequence is used as a testbed for the methodology; it is unique in that it contains 5,000 images, complex phenomena, and is feature rich. Oblique structures such as shock waves and shear layers are common in schlieren images. A vision-based methodology is used to provide an estimate of oblique structure angles through the unsteady sequence. The methodology has been applied to a complex flowfield with multiple shocks. A converged detection success rate between 94% and 97% for these structures is obtained. The modified curvature scale space is used to define features at salient points on shock contours. A challenge in developing methods for feature extraction in schlieren images is the reconciliation of existing techniques with features of interest to an aerodynamicist. Domain-specific knowledge of physics must therefore be incorporated into the definition and detec- tion phases. Known location and physically possible structure representations form a knowledge base that provides a unique feature definition and extraction. Model tip location and the motion of a shock intersection across several thousand frames are identified, localized, and tracked. Images are parsed into physically meaningful labels using segmentation. Using this representation, it is shown that in the double-cone flowfield, the dominant unsteady motion is associated with large scale random events within the aft-cone bow shock. Small scale organized motion is associated with the shock-separated flow on the fore-cone surface. We show that computer vision is a natural and useful extension to the evaluation of schlieren data, and that segmentation has the potential to permit new large scale measurements of flow motion.Item CAMERA SPECTRAL SENSITIVITY CHARACTERIZATION USING A BLACKBODY SOURCE(2011) Bedarkar, Rucha Sanjay; Sunderland, Peter B.; Mechanical Engineering; Digital Repository at the University of Maryland; University of Maryland (College Park, Md.)With digital cameras emerging as more effective tools for scientific research, there is increasing need for accurate and inexpensive ways to calibrate them. In particular, to date there has been no simple method to measure camera sensitivity as a function of wavelength. For example, narrow bandwidth monochromator beams are expensive and have calibration problems, while color chart method is unreliable owing to illumination dependence. This thesis presents a novel technique for spectral sensitivity calibration of a camera (or any black-and-white cameras or color sensors) using blackbody furnace operating at 650 - 1250 °C. Images recorded at 11 different temperatures are observed for red, green, and blue camera outputs. Using Planck ’ s Law to calculate the incident light intensities, the three color sensitivities as functions of wavelength are computed using MATLAB function that optimizes the spectral sensitivities until the blackbody measurements are closely matched. The results are in reasonable agreement with published sensitivities.Item Automated quantification and classification of human kidney microstructures obtained by optical coherence tomography(2009) Li, Qian; Chen, Yu; Bioengineering; Digital Repository at the University of Maryland; University of Maryland (College Park, Md.)Optical coherence tomography (OCT) is a rapidly emerging imaging modality that can non-invasively provide cross-sectional, high-resolution images of tissue morphology such as kidney in situ and in real-time. Because the viability of a donor kidney is closely correlated with its tubular morphology, and a large amount of image datasets are expected when using OCT to scan the entire kidney, it is necessary to develop automated image analysis methods to quantify the spatially-resolved morphometric parameters such as tubular diameter, and to classify various microstructures. In this study, we imaged the human kidney in vitro, quantified the diameters of hollow structures such as blood vessels and uriniferous tubules, and classified those structures automatically. The quantification accuracy was validated. This work can enable studies to determine the clinical utility of OCT for kidney imaging, as well as studies to evaluate kidney morphology as a biomarker for assessing kidney's viability prior to transplantation.Item INTEGRATED INPUT MODELING AND MEMORY MANAGEMENT FOR IMAGE PROCESSING APPLICATIONS(2005-12-07) Haim, Fiorella; Bhattacharyya, Shuvra S; Electrical Engineering; Digital Repository at the University of Maryland; University of Maryland (College Park, Md.)Image processing applications often demand powerful calculations and real-time performance with low power and energy consumption. Programmable hardware provides inherent parallelism and flexibility making it a good implementation choice for this application domain. In this work we introduce a new modeling technique combining Cyclo-Static Dataflow (CSDF) base model semantics and Homogeneous Parameterized Dataflow (HPDF) meta-modeling framework, which exposes more levels of parallelism than previous models and can be used to reduce buffer sizes. We model two different applications and show how we can achieve efficient scheduling and memory organization, which is crucial for this application domain, since large amounts of data are processed, and storing intermediate results usually requires the use of off-chip resources, causing slower data access and higher power consumption. We also designed a reusable wishbone compliant memory controller module that can be used to access the Xilinx Multimedia Board's memory chips using single accesses or burst mode.