Electrical and Computer Engineering Commons^™

Performance Analysis Of Video Coding For Machines With Vision Transformers, Vaishnavi Dhulipudi

Electronic Theses and Dissertations

This thesis investigates the performance of Video Coding for Machines (VCM) with Vision Transformer based object detection models. While existing VCM studies and tool designs have largely been developed under CNN-based assumptions, recent advances in computer vision have shown the growing importance of transformer based models. Motivated by this shift, this work studies whether VCM compressed data remains suitable for Vision Transformer based inference in addition to conventional CNN-based task networks.

To address this problem, three representative transformer based object detection models were selected: DETR, SWIN, and YOLOS. These models were chosen to represent different architectural styles, namely a CNN …

Spad Camera Image Analysis, Pratheen Reddy Pininti

Electronic Theses and Dissertations

I present a thorough noise characterization of the Canon MS-500, a Single-Photon Avalanche Diode (SPAD) camera system, tested under both lit and dark conditions. The camera outputs 10-bit digital number (DN) values produced by an internal processing pipeline whose design is not publicly documented. All analyses in this thesis therefore describe the camera’s DN output — the signal that any downstream detection, tracking, or classification system will actually receive — rather than the photon-counting statistics of the underlying SPAD array. All computations were performed on the native 10-bit data. Where a measured quantity has a known photon-counting analog, the relationship …

On Time-Series Analysis By Structured Matrix Decompositions With Applications To Signal Direction-Of-Arrival Estimation, Georgios Ierotheos Orfanidis

Electronic Theses and Dissertations

Modern autonomous systems operating in highly dynamic, non-stationary environments require reliable inference from short, potentially corrupted time-series measurements, where conventional statistical methods relying on large-sample support and stationarity assumptions become fundamentally inapplicable. This dissertation develops a unified, model-free theoretical framework for time-series analysis, with a particular application to signal Direction-of-Arrival (DoA) estimation, grounded in structured matrix decompositions under both the L2-norm and L1-norm formulations.

We first approach the problem from a conventional viewpoint by carrying out standard matrix analysis directly on Hankel-structured representations of time-series data. In this context, we demonstrate that L1-norm decompositions of Hankel matrices offer strong resistance …

Efficient Intrusion Detection For Iomt: Integrating Machine Learning, Feature Selection, And Fuzzy Logic, Ghaida Mansour Balhareth

Electronic Theses and Dissertations

The internet of medical things (IoMT) has transformed healthcare by enabling real-time patient monitoring, remote diagnoses, and effective data exchange among connected medical devices and clinical systems. The increasing reliance on interconnected medical equipment has also intensified cybersecurity risks, as resource-constrained devices and wireless communication channels are vulnerable to attacks such as man-in-the-middle, spoofing, data injection, and ransomware. Intrusion Detection Systems (IDSs) play a critical role in mitigating these threats; however, traditional IDS approaches often struggle with high-dimensional IoMT data, class imbalance, and uncertainty in traffic patterns, which can increase false alarms and reduce reliability in safety-critical environments. This dissertation …

Predictive Analytics In Oncology And Ophthalmology: Machine Learning Applications For Diabetic Retinopathy And Breast Cancer, Ali Abidalkareem

Electronic Theses and Dissertations

The convergence of artificial intelligence and healthcare represents one of the most transformative developments in modern medicine, with deep learning technologies emerging as powerful tools for addressing complex diagnostic challenges. This dissertation develops and validates machine learning frameworks that address critical challenges in medical diagnosis through innovative approaches to data augmentation, feature learning, and classification, focusing on two fundamental problems: Diabetic Retinopathy (DR) severity classification using multi-model convolutional neural networks (CNNs), and breast cancer stage identification using microRNA (miRNA) gene expression biomarkers. For diabetic retinopathy classification, this work proposes an ensemble deep learning framework that integrates Diffusion-based data augmentation for …

Graph-Based And Uncertainty-Aware Machine Learning For Academic Performance Prediction, Anca Olivia Muresan

Electronic Theses and Dissertations

Early identification of students at risk of academic failure is essential for timely pedagogical interventions and reducing dropout rates. While Artificial Intelligence (AI) has significantly advanced predictive modeling in education, two primary challenges persist: effectively modeling the complex, evolving relationships within heterogeneous educational data, and ensuring the reliability of model outputs for high-stakes decision-making. This dissertation addresses these challenges by proposing a comprehensive framework for early and continuous student performance prediction applied to the Open University Learning Analytics (OULA) dataset. First, we introduce a Heterogeneous Graph Neural Network (HGNN) approach that utilizes metapath structures to capture latent interactions between diverse …

Uas Path Planning With Dynamic Rerouting Using A Space-Time Graph, Kimoy Williams

Electronic Theses and Dissertations

UAS systems have emerged as multifaceted technologies with applications across a wide range of sectors. Their ability to access areas that are difficult or unsafe for manned systems has made them invaluable tools in various domains. As a result, UAS have transformed numerous industries, including infrastructure inspection, delivery and logistics, military and defense, as well as precision agriculture and environmental monitoring.

The advancement in UAS technology is fundamentally reliant upon ongoing research efforts in the specialized area of UAS path planning. Optimal flight planning is essential for a UAV to effectively execute its mission’s task safely, effectively, and in congruence …

Towards Optimal And Resilient Ac/Dc Microgrids: Control Design, Analysis, And Implementation, Jun Zhang

Electronic Theses and Dissertations

Microgrids serve as a small-scale power grid for utilizing renewable energy to enhance energy reliability, sustainability, and resilience. As an autonomous system, an islanded microgrid can disconnect from the utility grid and operate independently by maintaining system voltage and frequency. However, this new feature introduces coordination problems among distributed generators (DGs), such as 1) how to make sure the voltage profile and current sharing in DC microgrid with different types of converters; 2) how to reduce the impact of cyberattack when the system coordination is performed based on communication, and 3) how to calculate the steady state under a droop …

A Taxonomy-Driven Modular Defense Against Non-Canonical Language In Vision-Language-Action Models, Viraj Samson

Electronic Theses and Dissertations

Vision-Language-Action (VLA) models have recently achieved strong performance across manipulation benchmarks, but these benchmarks rely on highly templated instructions on which the models are typically fine-tuned, leaving their behavior under realistic language-side perturbation unclear. It remains an open question whether the linguistic flexibility inherited from vision-language pretraining survives this fine-tuning, or whether the resulting policies become narrowly tuned to benchmark phrasing and brittle to the intent-preserving language variation that real users naturally produce. We address this gap with a systematic study of VLA robustness under non-canonical instructions, comprising three components: a structured taxonomy of intent-preserving variations spanning linguistic, orthographic, and …

Fast And Sustainable Video Anomaly Detection With Continual Learning, Preethi Amasa

Electronic Theses and Dissertations

Real-time video anomaly detection systems deployed in surveillance, healthcare, and industrial environments face continuous distribution shifts in lighting, viewpoint, and activity patterns. Existing models often experience performance degradation under these conditions and may suffer catastrophic forgetting when adapting to new environments. This thesis proposes RegiGrow, a parameter-efficient continual adaptation framework built on the Flashback retrieval pipeline. RegiGrow integrates Mixture-of-Experts Low-Rank Adaptation into a frozen ImageBind encoder, enabling sequential domain adaptation without modifying the pretrained backbone. A lightweight router maps visual regime features to a distribution over LoRA experts, each specializing in a distinct normal operating regime. The central contribution is …

Data-Driven Optimization Of Memory Effects And Critical Components In Cascading Failure Interaction Networks, Md Farhan Tanvir

Electronic Theses and Dissertations

Cascading failures are a major concern for modern power systems‚ where a small failure can cascade through the network to create a large blackout. Because such an event could have catastrophic economic and social costs it is important to understand cascading failures‚ how to model them‚ and the possibility of reducing them. In this thesis‚ we develop a data-driven framework based on actual utility outage data to model and reduce cascading failures. The proposed methodology is based on generation-dependent interaction models and in this study interaction matrices obtained from historical outage events are used to describe the interaction between generations. …

Bridging Language And Game Worlds: Semantic Representations And Text-Driven Terrain Generation For Procedural Content, Zhongyu Xie

Electronic Theses and Dissertations

Procedural Content Generation (PCG) systems produce vast quantities of game levels, terrain, and environments, but lack semantic interfaces: no shared vocabulary exists between natural language, designer intent, and the structured representations generators operate on. This thesis addresses the language-content grounding gap in PCG through two complementary studies spanning semantic analysis and semantic synthesis. The first study introduces a group-supervised contrastive learning framework for semantic representation of symbolic PCG maps under many-to-one semantics, where visually distinct maps may share the same design intent. The framework combines parameter-guided semantic grouping, LLM-based caption augmentation, and a multi-positive contrastive objective that aligns language with …

High Frequency Dc-Dc Converter Design Optimization, Modeling And Control Based-Wbg Technology., Salah Ahmed Abdullah Eltief

Electronic Theses and Dissertations

The performance of DC-DC power converters is a cornerstone of modern electric vehicle (EV) powertrains, directly governing overall system efficiency, size, cost, and reliability. This dissertation presents a comprehensive performance analysis and optimization of DC-DC converter topologies to determine the most suitable design for high voltage EV applications. The evaluation rigorously compares efficiency, power losses, and physical size under a range of harsh operating conditions. A primary objective is to leverage Wide Bandgap (WBG) semiconductors, specifically Silicon Carbide (SiC), to push the performance boundaries of power conversion. While SiC devices are known for their superior material properties, a clear understanding …

Capacity, Allocation And Update Dynamics Of Human Memory Systems, Shaoying Wang

Electronic Theses and Dissertations

Information is encoded and stored in three types of memory: sensory memory (SM), short-term memory (STM), and long-term memory (LTM). SM has a large capacity but retains information for only a brief period. When information transfers to STM, only a limited amount can be stored. Information in STM can then be transferred to LTM, which has a much larger capacity and longer retention time. STM is often conceptualized as working memory (WM) to highlight its role in active information processing. Due to the limited capacity of STM, it is commonly believed that STM serves as the bottleneck for information processing. …

A Deep Learning Approach For Semantic Segmentation And Its Application On Ctc., Samir Farag Harb

Electronic Theses and Dissertations

This dissertation explores the modeling and analysis of medical images, focusing on the intricate task of colon segmentation and subsequent 3D reconstruction, which are critical steps in Computed Tomography Colonography (CTC) systems. The primary objective of this research is to develop precise segmentation approaches to enhance the accuracy of colon identification and reconstruction from abdominal CT scans. Three distinct segmentation approaches are proposed and evaluated: a Markov Random Field (MRF)-based approach, a convolutional neural network (CNN)-based deep learning (DL) approach, and a sequential episodic training with dual contrastive learning Approach (G-SET-DCL) that has a flavor of few-shot learning (FSL). To …

Development Of An Embedded Iot Board For Real-Time Floor Estimation Of Autonomous Robots, Carter J. Sorensen

Electronic Theses and Dissertations

As service robots become more prevalent in multi-story environments such as hospitals, hotels, and laboratories, accurate floor-level detection is critical to ensuring operational reliability. Consider a robot tasked with delivering medical samples in a multi-story laboratory. Without accurate feedback, a robot exiting on the wrong floor could introduce delays, disrupt workflows, or compromise sample integrity. Internet of Things (IoT) technologies offer a way to address these risks by providing real-time error detection and corrective capability. However, current IoT-based floor estimation systems often require invasive modifications to building infrastructure—particularly elevator control panels. These approaches introduce challenges related to cost, liability, backward …

An Advanced Hexacopter For Autonomous Exploration Of Mars: Attitude Control And Navigation Strategies, Laura Sopegno

Electronic Theses and Dissertations

Mars exploration has recently witnessed major interest within the scientific community. Unmanned robotic platforms offer reliable solutions to acquire and collect data and information from the Red Planet. Particularly, rovers, landers, and orbiters have significantly shaped planetary exploration on the Moon and Mars, contributing significantly to past missions while also highlighting limitations in their capacity to cover diverse terrains over wide ranges. Given current advances in Unmanned Aircraft Systems (UASs), Unmanned Aerial Vehicles (UAVs) offer promising alternatives for future scientific missions.

It is argued that hexacopters, with their relatively compact design and redundancy, present a promising …

Deep Learning In Lung Cancer Pre- And Post-Radiation Therapy: Diagnosis Of Malignancy And Radiation-Induced Lung Injury From 3d X-Ray Ct., Benjamin Peter Veasey

Electronic Theses and Dissertations

Lung cancer remains the leading cause of cancer-related mortality worldwide, with early detection and accurate diagnosis being critical for improving patient outcomes. Additionally, the progression of Radiation-Induced Lung Injury (RILI) following Stereotactic Body Radiation Therapy (SBRT) for lung cancer presents a significant diagnostic challenge. This dissertation addresses these challenges by developing deep learning-based diagnostic tools for both pre-treatment lung nodule malignancy classification and post-treatment RILI identification using 3D X-ray CT imaging. The research is divided into two primary objectives. First, for lung nodule malignancy classification, we developed a biopsy-confirmed dataset, called NLSTx, to train and evaluate deep learning models while …

Development Of A Portable Electroencephalogram With Display, Seth D. Davis

Electronic Theses and Dissertations

In this experiment, the goal was to design, build, and test an Electroencephalogram (EEG) circuit that would then be paired with a touchscreen display as a portable unit for use in Schizophrenic research at East Tennessee State University at the Quillen College of Medicine. The requirements for this prototype denoted that the device had to be able to obtain a signal from two electrodes placed on the head of an individual with a third attached to the ankle as a ground, amplify and filter the signal, and display the results on a touchscreen display for recording and analyzing. The circuit …

Multimodal Emotion Recognition For Human-Robot Interaction Across Neuro-Diverse Populations., Ruchik Mishra

Electronic Theses and Dissertations

This dissertation explores the integration of multimodal data streams and artificial intelligence pipelines to understand human affect in neurotypical and children with Autism Spectrum Disorder (ASD). This dissertation captures human affect in the context of human-robot interaction. For this, multiple studies have been presented with both children with ASD and neurotypical adults. This dissertation makes four contributions: 1) The first study introduces autonomy during perspective-taking teaching sessions by making verbal content generation through large language models (LLMs). This system is the first of its kind for teaching perspective-taking in a semi-autonomous manner under the supervision of domain experts. Furthermore, this …

An Automatic Colorectal Polyps Detection Approach For Ct Colonography., Mohamed Yousuf

Electronic Theses and Dissertations

Colon cancer, also known as colorectal cancer, is a significant health concern, with increasing incidence rates, particularly among individuals under 50. This rise has led experts to recommend the introduction of regular screenings at 45 years of age for adults at average risk. Early detection through such screenings can identify precancerous polyps, allowing their removal before they develop into cancer. This proactive approach has the potential to reduce colorectal cancer deaths by up to 60%. In addition, research indicates that people diagnosed before age 50 have better survival rates, which emphasizes the importance of early diagnosis. Therefore, adhering to recommended …

A Framework For Fair And Trustworthy Multimodal Video Anomaly Detection: Fusion, Dataset Quality Auditing, Explainability, And Fairness, Omeshamisu Judith Anigala

Electronic Theses and Dissertations

No abstract provided.

St-Hybrid: Dynamic Graph Learning With Multi-Scale Spatio-Temporal Attention For Traffic Forecasting, Dhe Yeong Ewaza Tchalla

Electronic Theses and Dissertations

Accurate short-term traffic forecasting is central to modern Intelligent Transportation Systems, supporting route guidance, adaptive signal control, and incident response. Yet producing reliable predictions remains difficult because traffic is highly non-stationary. The relationships among roadway sensors shift during congestion, incidents, weather changes, or fluctuations in demand, and the temporal structure of traffic spans several scales from abrupt minute-level variations to broader daily and weekly rhythms. Models that rely on fixed spatial graphs or a single temporal scale tend to miss these evolving and layered dependencies. This thesis addresses these challenges by developing a graph-learning framework that adapts to changing traffic …

Computation-Efficient Deep Learning Models For Computer Vision And Multimodal Vision-Language Tasks Via Network Pruning, Abir Mohammad Hadi

Electronic Theses and Dissertations

With the rapid evolution of deep neural networks over the past decade, the demand for efficient, generalizable, and task-adaptable models, especially in computer vision, has increased significantly. To address the computational and deployment challenges posed by overparameterized models, the research community has extensively explored model compression techniques such as pruning, quantization, and distillation. These approaches aim to enhance model efficiency without compromising performance, particularly when adapting to domain-specific tasks under limited resources. This dissertation investigates several underexplored yet critical aspects of task-aware deep learning model compression, spanning both convolutional and vision-language architectures. In the early part of this work, we …

Video Comprehension Score (Vcs): A Metric For Long-Form Video Description Evaluation, Harsh Dubey

Electronic Theses and Dissertations

Existing video description evaluation metrics fail to capture the long-range chronology and semantic alignment essential for long-form descriptions. An effective evaluation metric for long-form descriptions must (i) assess global thematic alignment, (ii) measure local semantic alignment, and (iii) evaluate chronological alignment while detecting corrupted content. We introduce Video Comprehension Score (VCS), a reference-based metric, which directly addresses these evaluation requirements through three components: Global Alignment Score for thematic alignment, Local Alignment Score for local semantic alignment, and Narrative Alignment Score for chronological alignment with adjustable tolerance. We evaluate VCS on two large-scale synthetic datasets designed to test corruption detection and …

Global Spectral Clustering Of Temporally Stable Pixels For Epics Identification, Validation, And Hyperspectral Assignment For Satellite Calibration, Juliana Maria Fajardo Rueda

Electronic Theses and Dissertations

No abstract provided.

Refinement Of Trend-To-Trend Cross Calibration Total Uncertainties Utilizing Extended Pseudo Invariant Calibration Sites (Epics) Global Temporally Stable Target, Minura Samaranayake

Electronic Theses and Dissertations

Cross-calibration is an essential technique for calibrating Earth Observation satellite sensors, which involves taking nearly simultaneous images of a ground target to compare uncalibrated sensor to a well-calibrated reference sensor. This study introduces the hyperspectral Trend-to-Trend (T2T) cross-calibration technique utilizing EPICS Cluster 13 Global Temporally Stable (Cluster 13-GTS) as the calibration target, offering better temporal stability than previous targets used in T2T cross-calibration by an absolute difference of 0.4%, between coefficients of variation across all bands excluding CA band. A multispectral sensor-specific normalized hyperspectral profile was developed using the EO-1 Hyperion hyperspectral profile over Cluster 13-GTS to improve Spectral Band …

Improving K-Mean Clustering: A Comparative Study Of Parallelized Version Of Modified K-Mean Algorithm For Clustering Of Satellite Images, Yuv Raj Pant

Electronic Theses and Dissertations

Efficient clustering of high-dimensional satellite image datasets remains a critical challenge, particularly due to the computational demands of spectral distance calculations, random centroid initialization, and sensitivity to outliers in conventional K-Mean algorithms. This study presents a comprehensive comparative analysis of eight parallelized variants of the K-means algorithm, designed to enhance clustering efficiency and reduce computational burden for large-scale satellite image analysis. The proposed parallelized implementations incorporate optimized centroid initialization for better starting point selection, a Dynamic K-mean sharp method to detect the outlier to improve cluster robustness, and a Nearest-Neighbor Iteration Calculation Reduction method to minimize redundant computations. These enhancements …

Semantic Think-On-Graph (Semtog) : Enhancing Graphrag Through Semantic Community Detection, Sugam Mishra

Electronic Theses and Dissertations

Graph-based Retrieval-Augmented Generation (GraphRAG) enhances large language models (LLMs) by grounding their reasoning in structured knowledge graphs, making them more reliable for multi-hop reasoning and factual QA. A central mechanism in Think-on-Graph systems such as ToG[14] and FastToG[7] is community detection, which groups locally related nodes into compact subgraphs so that LLMs can reason over focused, information-rich neighborhoods instead of traversing the entire graph. However, these methods rely purely on structural connectivity, often scattering semantically related entities across different communities and weakening the evidence provided to the LLM. We propose Semantic Think-on-Graph (SemToG), a semantic-aware extension of FastToG[7] that integrates …

Distributed Energy Resources (Der) Capacity Estimation With Spatiotemporal Downscaling For Transmission And Distribution Coordination, Abhilasha Suvedi

Electronic Theses and Dissertations

The primary objective of this thesis is to accurately estimate the capacity of distributed energy resources (DERs) using the spatiotemporally downscaled local solar irradiance for their efficient integration into transmission-level grid operations. Accurate estimation of DER capacity is crucial for their enhanced integration into real-time electricity markets. Different methods of spatiotemporally downscaling the solar irradiance are presented in this thesis, followed by their integration into the 12 house distribution network – a low voltage residential distribution system. Grid support functions (GSFs) like Volt-Watt Control, Volt-VAR control, and Frequency-Watt Control are applied to effectively regulate the frequency and maintain the voltage …