Open Access. Powered by Scholars. Published by Universities.®
- Discipline
-
- Biomedical Engineering and Bioengineering (43)
- Electrical and Computer Engineering (23)
- Civil and Environmental Engineering (21)
- Mechanical Engineering (15)
- Bioelectrical and Neuroengineering (9)
-
- Biomedical Devices and Instrumentation (9)
- Environmental Engineering (9)
- Medicine and Health Sciences (6)
- Civil Engineering (5)
- Physical Sciences and Mathematics (5)
- Bioimaging and Biomedical Optics (4)
- Biomechanical Engineering (3)
- Computer Engineering (3)
- Computer Sciences (3)
- Electrical and Electronics (3)
- Environmental Microbiology and Microbial Ecology (3)
- Life Sciences (3)
- Microbiology (3)
- Signal Processing (3)
- Chemical Engineering (2)
- Electro-Mechanical Systems (2)
- Electronic Devices and Semiconductor Manufacturing (2)
- Engineering Science and Materials (2)
- Hardware Systems (2)
- Medical Sciences (2)
- Medical Specialties (2)
- Power and Energy (2)
- Social and Behavioral Sciences (2)
- Software Engineering (2)
- Keyword
-
- Stroke (9)
- FMRI (5)
- Anaerobic Digestion (4)
- Sensor (4)
- Balance (3)
-
- Bioaugmentation (3)
- EEG (3)
- Electroencephalography (3)
- Electrooxidation (3)
- Fatigue (3)
- Finite element analysis (3)
- Gait (3)
- MEMS (3)
- Motor control (3)
- Optimization (3)
- Rehabilitation (3)
- Sensorimotor (3)
- Anaerobic digestion (2)
- Computational fluid dynamics (2)
- Computer vision (2)
- Concrete (2)
- Connectivity (2)
- Coordination (2)
- Dynamics (2)
- Electric machines (2)
- Electrocoagulation (2)
- Estimation (2)
- FRP (2)
- Finite Element (2)
- Finite Element Analysis (2)
Articles 1 - 30 of 151
Full-Text Articles in Engineering
Closing The Phosphorus Loop With Green Infrastructure: Removal And Recovery Using Amended Soils And 3d-Printed Material, Colin Wilson
Closing The Phosphorus Loop With Green Infrastructure: Removal And Recovery Using Amended Soils And 3d-Printed Material, Colin Wilson
Dissertations (1934 -)
Phosphorus (P) is a life-essential element aiding the deterioration of water quality. Stormwater runoff transports nonpoint source P pollution to waterbodies, contributing to excessive P loading which may result in eutrophication. Green infrastructure (GI) is implemented to solve stormwater runoff pollution, but the P removal performance of these systems is variable. Moreover, P is arguably being unsustainably mined for fertilizer production and animal feed supplements to support demand from increasing global populations. This dissertation explored advancing the circular P economy utilizing additive manufacturing and GI soil amendments to remove and recover P from stormwater runoff. Laboratory-scale batch experiments were used …
Antibiotic Resistance In Urban Stormwater Runoff And Green Stormwater Infrastructure, Kassidy N. O'Malley
Antibiotic Resistance In Urban Stormwater Runoff And Green Stormwater Infrastructure, Kassidy N. O'Malley
Dissertations (1934 -)
Antibiotic resistance in the environment poses significant risks to human health, especially through the transmission of antibiotic resistance genes (ARGs) and antibiotic-resistant bacteria (ARB) via pathways like food and water. Urban stormwater runoff has been identified as a significant contributor to the release and transport of ARGs in the environment but has remained under-researched. This dissertation aimed to address this gap by investigating the fate of ARGs in urban stormwater runoff to guide future stormwater management decisions. Two phases were completed; the first sought to investigate extracellular antibiotic resistance genes (eARGs) in stormwater runoff, and the second focused on quantifying …
The Role Of Membrane Porosity In Ion Removal By Ro/Nf Membranes And Development Of A Regenerable Pac-Loaded Membrane, Yizhi Hou
Dissertations (1934 -)
Reverse osmosis (RO) and nanofiltration (NF) membranes are widely used for water treatment applications. They are charged porous materials, in which size exclusion and charge exclusion work together to dictate solute transport. In this study, membrane porosity was primarily determined by pore size. Thicker, more porous membranes with smaller pores had higher NaCl removal and less permeate water flux. Higher porosity membranes have greater inner pore surface area available to support a greater number of functional groups, thereby rejecting ion flux and increasing ion removal. The Donnan steric pore model (DSPM) was used to determine that the contribution of diffusion …
The Effect Of Microstructural Defects And Geometrical Features On Fatigue Behavior Of Superelastic Nitinol Wires, Parisa Shabani Nezhad
The Effect Of Microstructural Defects And Geometrical Features On Fatigue Behavior Of Superelastic Nitinol Wires, Parisa Shabani Nezhad
Dissertations (1934 -)
Nitinol is an alloy of nickel and titanium, which exhibits outstanding functional properties, such as shape memory and superelastic behavior. Superelastic nitinol wires exhibit recoverable strains that are significantly greater than traditional alloys. In most of its applications, nitinol is exposed to cyclic loads, which results in functional and/or structural fatigue that ultimately leads to failure. The classical fatigue theories do not appropriately address the fatigue performance of superelastic nitinol due to the complex nature of the martensitic transformation. It has been shown that the main fatigue crack initiation sites, other than surfaces, are microstructural inhomogeneities such as voids and …
Design And Advanced Model Predictive Control Of Wide Bandgap Based Power Converters, Waqar Ahmed Khan
Design And Advanced Model Predictive Control Of Wide Bandgap Based Power Converters, Waqar Ahmed Khan
Dissertations (1934 -)
The field of power electronics (PE) is experiencing a revolution by harnessing the superior technical characteristics of wide-band gap (WBG) materials, namely Silicone Carbide (SiC) and Gallium Nitride (GaN). Semiconductor devices devised using WBG materials enable high temperature operation at reduced footprint, offer higher blocking voltages, and operate at much higher switching frequencies compared to conventional Silicon (Si) based counterpart. These characteristics are highly desirable as they allow converter designs for challenging applications such as more-electric-aircraft (MEA), electric vehicle (EV) power train, and the like. This dissertation presents designs of a WBG based power converters for a 1 MW, 1 …
Electrocoagulation For Treatment Of Disinfection Byproducts, Organic Matter, And Per-And Polyfluoroalykl Substances., Donald Rockwood Ryan
Electrocoagulation For Treatment Of Disinfection Byproducts, Organic Matter, And Per-And Polyfluoroalykl Substances., Donald Rockwood Ryan
Dissertations (1934 -)
Technologies are needed to treat contaminants such as disinfection byproducts (DBPs), trace organic compounds (TOrCs), and per- and polyfluoroalkyl substances (PFAS) in water to improve consumer safety and mitigate chronic risk. This dissertation focused on evaluating electrocoagulation (EC) and hydrogen peroxide (H2O2)-enhanced EC (peroxi-electrocoagulation; EC:H2O2) for treatment of these contaminants. The impacts of source water quality (such as natural organic matter [NOM]) on overall performance was assessed and the non-destructive and destructive pathways involved in treatment were evaluated. The first objective focused on EC performance for mitigating the formation of regulated DBPs in water to better substantiate EC’s performance relative …
Downstream Task Self-Supervised Learning For Object Recognition And Tracking, Abubakar Siddique
Downstream Task Self-Supervised Learning For Object Recognition And Tracking, Abubakar Siddique
Dissertations (1934 -)
This dissertation addresses three limitations of deep learning methods in image and video understanding-based machine vision applications. Firstly, although deep convolutional neural networks (CNNs) are efficient for image recognition applications such as object detection and segmentation, they perform poorly under perspective distortions. In real-world applications, the camera perspective is a common problem that we can address by annotating large amounts of data, thus limiting the applicability of the deep learning models. Secondly, the typical approach for single-camera tracking problems is to use separate motion and appearance models, which are expensive in terms of computations and training data requirements. Finally, conventional …
Urban Water Quality: Socio-Economic Distribution Of Stream Degradation, And The Influence Of Climate On Bmp Performance, Isabelle Horvath
Urban Water Quality: Socio-Economic Distribution Of Stream Degradation, And The Influence Of Climate On Bmp Performance, Isabelle Horvath
Dissertations (1934 -)
Urban water quality impairments have long burdened urban aquatic ecosystems in a phenomenon termed “urban stream syndrome”. The symptoms of this “syndrome” include physical changes in stream morphology and water level, biologically stressed environments, and perturbations in ecosystem processes. Despite years of research and costly investments in restoration and rehabilitation, urban waterways are still plagued by degradation. Improvement in urban water quality will require multi-faceted efforts, including progress on 2 key fronts 1) increased understanding of current impairments, and 2) increased knowledge about urban stormwater infrastructure like best management practices (BMPs). Towards both topics, engineers have quantified the influence of …
Feasibility Of Sewer Overflow Treatment Using Chemical Oxidation After Rapid Solids Removal, Paige Peters
Feasibility Of Sewer Overflow Treatment Using Chemical Oxidation After Rapid Solids Removal, Paige Peters
Dissertations (1934 -)
Sewer overflows discharge 850 billion gallons of untreated sewage into U.S. lakes and rivers every year during high-intensity precipitation events. Common approaches include storage tanks or rapid solids removal followed by chlorine disinfection. These approaches can be capital intensive and only achieve partial treatment. A decentralized, end-of-pipe sewer overflow treatment system would eliminate detrimental overflow effects while handling highly variable flow and pollutant loading. Specifically, the rapid reaction rate of chemical oxidation can achieve soluble organics removal and disinfection of chlorine-resistant pathogens in a small footprint. To address the overflow issue, an advanced high-rate treatment system (AH-RTS) was developed and …
Evaluating The Impact Of Land Use Changes, Drivers Of Tmdl Development, And Green Infrastructure On Stream Impairments, Charitha Jayasri Gunawardana
Evaluating The Impact Of Land Use Changes, Drivers Of Tmdl Development, And Green Infrastructure On Stream Impairments, Charitha Jayasri Gunawardana
Dissertations (1934 -)
Despite the water quality improvements and regulatory advancements over the last 50 years since the enactment of the Clean Water Act, water bodies within the United States are still impaired for a broad range of contaminants from non-point source pollution. Improving watershed management approaches to meet this challenge will require a greater understanding of (1) how changes within a watershed, such as changing land use, impact stream water quality, (2) what influence socioeconomic, spatial and political factors may have on the progress towards meeting water quality goals, such as those set within Total Maximum Daily Loads (TMDLs), and (3) how …
A Study On The Hierarchical Control Structure Of The Islanded Microgrid, Fahad Shumran Alshammari
A Study On The Hierarchical Control Structure Of The Islanded Microgrid, Fahad Shumran Alshammari
Dissertations (1934 -)
The microgrid is essential in promoting the power system’s resilience through its ability to host small-scale DG units. Furthermore, the microgrid can isolate itself during main grid faults and supply its demands. However, islanded operation of the microgrid is challenging due to difficulties in frequency and voltage control. In islanded mode, grid-forming units collaborate to control the frequency and voltage. A hierarchical control structure employing the droop control technique provides these control objectives in three consecutive levels: primary, secondary, and tertiary. However, challenges associated with DG units in the vicinity of distribution networks limit the effectiveness of the islanded mode …
Continuous Myoelectric Prediction Of Future Ankle Kinematics And Kinetics For The Control Of Active Powered Prostheses, Erika Virginia Zabre González
Continuous Myoelectric Prediction Of Future Ankle Kinematics And Kinetics For The Control Of Active Powered Prostheses, Erika Virginia Zabre González
Dissertations (1934 -)
The purpose of this dissertation is to develop a continuous predictive model of gait that incorporates user intent, vis-à-vis surface electromyography (EMG) signals of the lower limb, to continuously predict future ankle kinematics and kinetics across multiple terrains and their transitions for the control of an active ankle-foot prosthesis. Continuously predicting variations in gait, particularly during transitions and noncyclic activities, is limited in lower limb prostheses. Despite performance improvements in active lower limb prostheses, there remains a need for robust and accurate control that incorporates direct user intent (e.g., myoelectric control) to limit physical and cognitive demands and provide more …
Improving Char Combustion Models Using Micro-Ct, Automated Image Analysis, And Pore-Resolving Simulations, Dongyu Liang
Improving Char Combustion Models Using Micro-Ct, Automated Image Analysis, And Pore-Resolving Simulations, Dongyu Liang
Dissertations (1934 -)
Combustion of pulverized coal and biomass in furnaces and boilers involves millions of particles with a distribution of sizes and morphologies. Because char particles typically react under zone II conditions, in which pore-diffusion and heterogenous reaction both influence the rate of conversion, the morphology of the porous particles affects reactor-scale outputs. To understand the impacts of complex morphology, 3-D pore-resolving simulations employing real char geometries obtained from high-resolution X-ray microtomography are first used to study combustion of 150 coal char particles and 30 biomass char particles. Localized reactant penetration into the innermost regions of the particles is observed, facilitated by …
Semantic Representation Of Road Infrastructure Information, Tianjiao Zhao
Semantic Representation Of Road Infrastructure Information, Tianjiao Zhao
Dissertations (1934 -)
A multi-sectoral collaboration completes a successful transportation infrastructure project. The cooperation involves designers, contractors, operators, users, government agencies, and maintenance staffs. Throughout the project's life cycle, a huge amount of data is generated and stored in various sectors. Therefore, an efficient information cross-sector exchange approach is necessary. Additionally, the World Wide Web is ubiquitous and enmeshed with multiple business processes. Therefore, it is imperative to represent business information in a format that improves information exchange as well as automated processing of business data. Ideally, road data scattered across different information sources, such as design software, geographic information systems (GIS), cost …
Improved Management Of Recalcitrant Nutrient Species: Transformation And Adsorption, Synthia Parveen Mallick
Improved Management Of Recalcitrant Nutrient Species: Transformation And Adsorption, Synthia Parveen Mallick
Dissertations (1934 -)
Soluble non-reactive nutrient species, i.e., dissolved organic nitrogen (DON) and soluble non-reactive phosphorus (sNRP), are not effectively removed and recovered. Unfortunately, the non-reactive species can cause eutrophication in receiving waterbodies. Thus, removal and recovery of soluble non-reactive nutrients is critical for reducing nutrient discharge and advancing the national goal of enhanced nutrient recovery. Transformation of non-reactive nutrients to more readily removable/recoverable species using ozonation and UV/H2O2 for enhanced nutrient recovery has been reported in literature. Electrooxidation (EO) may outperform these processes in transforming nutrients as EO can utilize multiple oxidation pathways, e.g., in-situ generated oxidants or direct electron transfer (DET). …
Immobilized Phosphate-Binding Protein Systems To Remove And Recover Phosphate, Faten Bakri Hussein
Immobilized Phosphate-Binding Protein Systems To Remove And Recover Phosphate, Faten Bakri Hussein
Dissertations (1934 -)
Inorganic phosphorus (Pi) is a critical element for all life and a major component of DNA’s backbone. Global agriculture requires Pi to support the rapid growth of plants and high crop productivity. Yet, the mineable Pi supply needed for global agriculture is in short supply in nature and Pi reserves are unevenly distributed around the world. In contrast, surplus Pi in water poses a serious pollution problem as it can cause eutrophication. Scarcity of non-renewable Pi resources combined with pollution in the form of overabundant Pi presents a phosphorus paradox. Tackling this issue is necessary and using innovative approaches targeting …
Novel Test Fixture For Characterizing Mems Switch Microcontact Reliability And Performance, Protap Kumar Mahanta
Novel Test Fixture For Characterizing Mems Switch Microcontact Reliability And Performance, Protap Kumar Mahanta
Dissertations (1934 -)
Microelectromechanical systems (MEMS) switch is considered as a better alternative than the conventional solid-state DC and RF switches due to their low contact resistance approximately 1 Ω, near-zero power consumption approximately 0 W, low insertion loss approximately 0.2 dB, and high isolation approximately -30 dB. However, reliability is a great concern for them to be ubiquitously used by the industry for specific applications. Switching dynamics and microcontact surface physics play the critical role in determining their reliability. A simple, quick, and efficient test fixture is required to study the contact surface physics as well as to optimize the switching dynamics.In …
Evaluation Of Shoulder Complex Kinematics Pre And Post Rotator Cuff Repair And Effects Of Altering Rotation Sequence, Alyssa Joy Schnorenberg
Evaluation Of Shoulder Complex Kinematics Pre And Post Rotator Cuff Repair And Effects Of Altering Rotation Sequence, Alyssa Joy Schnorenberg
Dissertations (1934 -)
Approximately 25% of adults in the United States will sustain a rotator cuff tear in their lifetime. Rotator cuff tears can impede one's ability to perform activities of daily living and maintain functional independence. Nearly 300,000 rotator cuff repair surgeries are performed annually and aim to decrease pain, increase range of motion, and enable return to the workforce.The goal of postoperative rehabilitation is to allow healing of the repaired cuff while restoring function and range of motion, but it is demanding and may require a year. However, most rehabilitation protocols are based solely on clinical experience rather than on evidence-based …
Eeg During Motor Tasks In Stroke: The Effects Of Remote Ischemic Conditioning And Fatigue On Brain Activity, Priya Balasubramanian
Eeg During Motor Tasks In Stroke: The Effects Of Remote Ischemic Conditioning And Fatigue On Brain Activity, Priya Balasubramanian
Dissertations (1934 -)
This dissertation aimed to use electroencephalography (EEG) to identify the effects of fatigue and remote ischemic conditioning on brain activity. Lesions due to stroke directly or indirectly affect regions of the brain and the descending corticospinal pathways. Cortical reorganization and alternate descending neural pathways are used during recovery from stroke as compensation mechanisms for motor deficits. These mechanisms exacerbate the deficits by worsening the ability to terminate muscle activity, individuate muscles for fine motor control and minimize abnormal muscle synergy and coactivation patterns to conserve resources during movement. Even though imaging and muscle activation studies have documented the existence and …
Impact Of Corrosion Inhibitors And Metals On Antibiotic Resistance In Drinking Water Distribution Systems, Lee Kimbell
Impact Of Corrosion Inhibitors And Metals On Antibiotic Resistance In Drinking Water Distribution Systems, Lee Kimbell
Dissertations (1934 -)
Drinking water distribution systems are important for transporting clean drinking water. Antibiotic resistant bacteria (ARB) and antibiotic resistance genes (ARGs) in drinking water may pose risks to human and environmental health. Aging infrastructure such as metal pipes can experience significant corrosion, which can have impacts on the chemical and microbial water quality inside of drinking water systems. Chemicals are commonly added to drinking water to prevent corrosion (i.e., corrosion inhibitors). The metal pipe materials and corrosion inhibitors containing metals may have impacts on the bacteria and ARGs in drinking water systems. This research investigated the presence of ARGs in real-world …
Micromechanics Of Additively Manufactured Materials Under Dynamic Loading, Christopher Rueben Johnson
Micromechanics Of Additively Manufactured Materials Under Dynamic Loading, Christopher Rueben Johnson
Dissertations (1934 -)
Ongoing works aim to develop a predictive computational framework whichcan be utilized to shorten the design and certification cycle of additively manufactured (AM) 304L stainless steel (SS304L). High fidelity experiments are utilized to validate and parameterize the computational framework. Results have detailed differing mechanical responses across a breadth of strain rates for wrought and AM SS304L. Differences are attributed to the microstructural properties resulting from the fabrication and treatment of the materials. This work aims to supplement preexisting high strain rate compressive strength experimental data sets to further quantify process-structure-property (PSP) relationships in development of the computational framework. Materials investigated …
Deep Convolutional Correlation Particle Filter For Visual Tracking, Reza Jalil Mozhdehi
Deep Convolutional Correlation Particle Filter For Visual Tracking, Reza Jalil Mozhdehi
Dissertations (1934 -)
In this dissertation, we explore the advantages and limitations of the application of sequential Monte Carlo methods to visual tracking, which is a challenging computer vision problem. We propose six visual tracking models, each of which integrates a particle filter, a deep convolutional neural network, and a correlation filter. In our first model, we generate an image patch corresponding to each particle and use a convolutional neural network (CNN) to extract features from the corresponding image region. A correlation filter then computes the correlation response maps corresponding to these features, which are used to determine the particle weights and estimate …
Direct And Non-Invasive Monitoring Of Battery Internal State Via Novel Gmi-Idt Magnetic Sensor, Akila Khatun
Direct And Non-Invasive Monitoring Of Battery Internal State Via Novel Gmi-Idt Magnetic Sensor, Akila Khatun
Dissertations (1934 -)
Efficient battery management systems (BMSs) in rechargeable battery-based systems require precise measurements of various battery parameters including state of charge (SOC), state of health (SOH) and charge capacity. Presently, SOC, charge capacity and SOH can only be indirectly inferred from long-term measurement of current, open circuit voltage (OCV), and temperature using multiple sensors. These techniques can only give an approximation of SOC and often require knowledge of the recent battery history to prevent excessive inaccuracy.To improve the performance of the BMS, an alternative method of monitoring the internal state of Li-ion batteries is presented here. Theoretical analysis of Li-ion batteries …
Dynamic Material Properties Copper Mountain Sandstone Under Uniaxial And Oblique Impact, Nathaniel Steven Helminiak
Dynamic Material Properties Copper Mountain Sandstone Under Uniaxial And Oblique Impact, Nathaniel Steven Helminiak
Dissertations (1934 -)
A generalized P-Lambda equation of state and strength model (dependent on hydrostatic and deviatoric stress terms) for porous alpha phase crystalline quartz under compression load was generated utilizing Copper Mountain sandstone (from Mosinee WI) and augmented with existing petrographic data. Tests with Copper Mountain sandstone samples began with an examination of elastic properties before conventional low strain rate testing was conducted. Sample thickness varying from ~0.5 mm to ~3.0 mm, were tested in high strain rate plane strain and pressure-shear configurations at impact velocities from 3 m/s to 720 m/s. Each plane strain test contained multiple sandstone samples so that …
Thermal Deformation In High Pressure Die Casting Shot Sleeves, Ross Crowley
Thermal Deformation In High Pressure Die Casting Shot Sleeves, Ross Crowley
Dissertations (1934 -)
Replacement of shot sleeves represents a significant cost in cold chamber high pressure die casting (HPDC) processes. It is well known that the service life of a shot sleeve is closely related to its thermomechanical deformation behavior, so an improved understanding of this behavior is needed to guide the design of shot sleeves for increased service life. Prior research efforts have analyzed shot sleeve deformation but were based on 2D representations and did not consider transient effects from the molten metal’s flow behavior. This objective of this study was to analyze thermo-mechanical deformation during the injection sequence by using a …
Design Of Novel Experiments And Analyses For Head And Spine Trauma Biomechanics, John Robert Humm
Design Of Novel Experiments And Analyses For Head And Spine Trauma Biomechanics, John Robert Humm
Dissertations (1934 -)
Previous biomechanics research studies have used both whole-body and isolated postmortem human surrogate experiments to define human injury tolerances, advance safety in injury producing environments, and promulgate standards for design of injury mitigating systems. Recent developments in transportation and sports-related fields have led to an increasing need to determine tolerances for combined loading (multi-axis) scenarios. This dissertation demonstrates the efficacy of the novel experimental design and analysis to head and spine trauma in these modalities. The first topic was the design of a novel experiment to examine the effect oblique loading on the tension tolerance of the lumbar spine. To …
Study Of Techniques To Enhance The Performance Of Concrete Members Strengthened By Externally Bonded Composite Materials, Zhao Wang
Dissertations (1934 -)
External bonding of fiber reinforced polymer (FRP) composite materials has been widely recognized as a popular technique for strengthening concrete structures. To find advanced strategies to improve the bond behavior of externally bonded composites strengthened concrete, three major tasks were conducted in this study: experimental work, finite element (FE) analysis, and reliability analysis. The mechanical properties and the deterioration due to water absorption of carbon nanofibers (CNFs) epoxy composites with different contents of CNFs were tested through uniaxial tensile test. According to the tensile test results, adding CNFs could significantly enhance the tensile behavior for epoxy resin. Because adding proper …
Deep Learning For Quantitative Susceptibility Mapping Reconstruction, Juan Liu
Deep Learning For Quantitative Susceptibility Mapping Reconstruction, Juan Liu
Dissertations (1934 -)
Quantitative susceptibility mapping (QSM) is a magnetic resonance imaging (MRI) technique that estimates tissue magnetic susceptibility from Larmor frequency offset measurements. The generation of QSM requires solving ill-posed background field removal (BFR) and field-to-source inversion problems. Incorrect BFR often introduces erroneous local field outputs and subsequently affects susceptibility quantification accuracy. Inaccurate field-to-source inversion often causes large susceptibility estimation errors that appear as streaking artifacts in the QSM, especially in massive hemorrhagic regions. Because current QSM techniques struggle to generate reliable QSM, the clinical translation of QSM is greatly hindered. Recently, deep learning (DL) has achieved state-of-the-art performance in many computer …
Overcoming Anaerobic Digestion Toxicity Of Aqueous Liquid From Wastewater Solids Pyrolysis, Seyedehfatemeh Seyedi
Overcoming Anaerobic Digestion Toxicity Of Aqueous Liquid From Wastewater Solids Pyrolysis, Seyedehfatemeh Seyedi
Dissertations (1934 -)
Pyrolysis treats and potentially recovers energy from wastewater solids (WWS). Aqueous pyrolysis liquid (APL), however, is produced and its management is a bottleneck to pyrolysis full-scale application. To overcome this bottleneck, anaerobic digestion (AD) may be a possible method to manage APL, but APL digestion has not been conclusively demonstrated. In AD, a select group of microorganisms convert organic chemicals to biogas containing methane that can be used as a fuel. In this dissertation work, APL derived from WWS pyrolysis was successfully converted into methane as the sole substrate and as a co-digestate with synthetic primary sludge in long-term, continuous …
Stochastic Methods For Fine-Grained Image Segmentation And Uncertainty Estimation In Computer Vision, Philipe Ambrozio Dias
Stochastic Methods For Fine-Grained Image Segmentation And Uncertainty Estimation In Computer Vision, Philipe Ambrozio Dias
Dissertations (1934 -)
In this dissertation, we exploit concepts of probability theory, stochastic methods and machine learning to address three existing limitations of deep learning-based models for image understanding. First, although convolutional neural networks (CNN) have substantially improved the state of the art in image understanding, conventional CNNs provide segmentation masks that poorly adhere to object boundaries, a critical limitation for many potential applications. Second, training deep learning models requires large amounts of carefully selected and annotated data, but large-scale annotation of image segmentation datasets is often prohibitively expensive. And third, conventional deep learning models also lack the capability of uncertainty estimation, which …