Engineering Commons^™

Accelerating Machine Learning Inference For Satellite Component Feature Extraction Using Fpgas., Andrew Ekblad

Theses and Dissertations

Running computer vision algorithms requires complex devices with lots of computing power, these types of devices are not well suited for space deployment. The harsh radiation environment and limited power budgets have hindered the ability of running advanced computer vision algorithms in space. This problem makes running an on-orbit servicing detection algorithm very difficult. This work proposes using a low powered FPGA to accelerate the computer vision algorithms that enable satellite component feature extraction. This work uses AMD/Xilinx’s Zynq SoC and DPU IP to run model inference. Experiments in this work centered around improving model post processing by creating implementations …

Developing A Flexible System For A Friendly Robot To Ease Dementia (Fred) Using Cloud Technologies And Software Design Patterns, Robert James Bray

Masters Theses

In this work, we designed two prototypes for a friendly robot to ease dementia (FRED). This affordable social robot is designed to provide company to older adults with cognitive decline, create reminders for important events and tasks, like taking medication, and providing cognitive stimulus through games. This project combines several cloud technologies including speech-to-text, cloud data storage, and chat generation in order to provide high level interactions with a social robot. Software design patterns were employed in the creation of the software to produce flexible code base that can sustain platform changes easily, including the framework used for the graphical …

Cybersecurity In Critical Infrastructure Systems: Emulated Protection Relay, Mitchell Bylak

Computer Science and Computer Engineering Undergraduate Honors Theses

Cyber-attacks on Critical Systems Infrastructure have been steadily increasing across the world as the capabilities of and reliance on technology have grown throughout the 21st century, and despite the influx of new cybersecurity practices and technologies, the industry faces challenges in its cooperation between the government that regulates law practices and the private sector that owns and operates critical infrastructure and security, which has directly led to an absence of eas- ily accessible information and learning resources on cybersecurity for use in public environments and educational settings. This honors research thesis addresses these challenges by submitting the development of an …

Deep Learning Frameworks For Accelerated Magnetic Resonance Image Reconstruction Without Ground Truths, Ibsa Kumara Jalata

Graduate Theses and Dissertations

Magnetic Resonance Imaging (MRI) is typically a slow process because of its sequential data acquisition. To speed up this process, MR acquisition is often accelerated by undersampling k-space signals and solving an ill-posed problem through a constrained optimization process. Image reconstruction from under-sampled data is posed as an inverse problem in traditional model-based learning paradigms. While traditional methods use image priors as constraints, modern deep learning methods use supervised learning with ground truth images to learn image features and priors. However, in some cases, ground truth images are not available, making supervised learning impractical. Recent data-centric learning frameworks such as …

Quiz Web Application, Dipti Rathod

Electronic Theses, Projects, and Dissertations

The Quiz web application is designed to facilitate the process of quiz creation and participation. This web application mainly consists of three roles: Admin, Instructor, and Student. Each role has specific features, functionalities, and permissions. With a user-friendly interface, the admin role can handle the departments, courses, and instructors. This web application also ensures smooth quiz management, allowing the instructors to schedule the upcoming quizzes, create the questions, and manage the students with ease. Student roles have features like taking quizzes and seeing their results. Additionally, this web application includes a significant feature to prevent cheating during online tests, ensuring …

Towards Multi-Modal Interpretable Video Understanding, Quang Sang Truong

Graduate Theses and Dissertations

This thesis introduces an innovative approach to video comprehension, which simulates human perceptual mechanisms and establishes a comprehensible and coherent narrative representation of video content. At the core of this approach lies the creation of a Visual-Linguistic (VL) feature for an interpretable video portrayal and an adaptive attention mechanism (AAM) aimed at concentrating solely on principal actors or pertinent objects while modeling their interconnections. Taking cues from the way humans disassemble scenes into visual and non-visual constituents, the proposed VL feature characterizes a scene via three distinct modalities: (i) a global visual environment, providing a broad contextual comprehension of the …

Real-Time Analysis Of Aerosol Size Distributions With The Fast Integrated Mobility Spectrometer (Fims), Daisy Wang

McKelvey School of Engineering Theses & Dissertations

The Fast Integrated Mobility Spectrometer (FIMS) has emerged as an innovative instrument in the aerosol science domain. It employs a spatially varying electric field to separate charged aerosol particles by their electrical mobilities. These separated particles are then enlarged through vapor condensation and imaged in real time by a high-speed CCD camera. FIMS achieves near 100% detection efficiency for particles ranging from 10 nm to 600 nm with a temporal resolution of one second. However, FIMS’ real-time capabilities are limited by an offline data analysis process. Deferring analysis until hours or days after measurement makes FIMS' capabilities less valuable for …

Assessing Blockchain’S Potential To Ensure Data Integrity And Security For Ai And Machine Learning Applications, Aiasha Siddika

Master of Science in Information Technology Theses

The increasing use of data-centric approaches in the fields of Machine Learning and Artificial Intelligence (ML/AI) has raised substantial issues over the security, integrity, and trustworthiness of data. In response to this challenge, Blockchain technology offered a promising and practical solution, as its inherent characteristics as a decentralized distributed ledger, coupled with cryptographic processes, offer an unprecedented level of data confidentiality and immutability. This study examines the mutually beneficial connection between Blockchain technology and ML/AI, using Blockchain's inherent capacity to protect against unauthorized alterations of data during the training phase of ML models. The method involves building valid blocks of …

Ensuring Non-Repudiation In Long-Distance Constrained Devices, Ethan Blum

Undergraduate Honors Theses

Satellite communication is essential for the exploration and study of space. Satellites allow communications with many devices and systems residing in space and on the surface of celestial bodies from ground stations on Earth. However, with the rise of Ground Station as a Service (GsaaS), the ability to efficiently send action commands to distant satellites must ensure non-repudiation such that an attacker is unable to send malicious commands to distant satellites. Distant satellites are also constrained devices and rely on limited power, meaning security on these devices is minimal. Therefore, this study attempted to propose a novel algorithm to allow …

Brunet: Disruption-Tolerant Tcp And Decentralized Wi-Fi For Small Systems Of Vehicles, Nicholas Brunet

Master's Theses

Reliable wireless communication is essential for small systems of vehicles. However, for small-scale robotics projects where communication is not the primary goal, programmers frequently choose to use TCP with Wi-Fi because of their familiarity with the sockets API and the widespread availability of Wi-Fi hardware. However, neither of these technologies are suitable in their default configurations for highly mobile vehicles that experience frequent, extended disruptions. BRUNET (BRUNET Really Useful NETwork) provides a two-tier software solution that enhances the communication capabilities for Linux-based systems. An ad-hoc Wi-Fi network permits decentralized peer-to-peer and multi-hop connectivity without the need for dedicated network infrastructure. …

Early-Warning Prediction For Machine Failures In Automated Industries Using Advanced Machine Learning Techniques, Satnam Singh

Electronic Theses, Projects, and Dissertations

This Culminating Experience Project explores the use of machine learning algorithms to detect machine failure. The research questions are: Q1) How does the quality of input data, including issues such as outliers, and noise, impact the accuracy and reliability of machine failure prediction models in industrial settings? Q2) How does the integration of SMOTE with feature engineering techniques influence the overall performance of machine learning models in detecting and preventing machine failures? Q3) What is the performance of different machine learning algorithms in predicting machine failures, and which algorithm is the most effective? The research findings are: Q1) Effective outlier …

Improving Credit Card Fraud Detection Using Transfer Learning And Data Resampling Techniques, Charmaine Eunice Mena Vinarta

Electronic Theses, Projects, and Dissertations

This Culminating Experience Project explores the use of machine learning algorithms to detect credit card fraud. The research questions are: Q1. What cross-domain techniques developed in other domains can be effectively adapted and applied to mitigate or eliminate credit card fraud, and how do these techniques compare in terms of fraud detection accuracy and efficiency? Q2. To what extent do synthetic data generation methods effectively mitigate the challenges posed by imbalanced datasets in credit card fraud detection, and how do these methods impact classification performance? Q3. To what extent can the combination of transfer learning and innovative data resampling techniques …

Detection Of Myofascial Trigger Points With Ultrasound Imaging And Machine Learning, Benjamin Formby

All Theses

Myofascial Pain Syndrome (MPS) is a common chronic muscle pain disorder that affects a large portion of the global population, seen in 85-93% of patients in specialty pain clinics [10]. MPS is characterized by hard, palpable nodules caused by a stiffened taut band of muscle fibers. These nodules are referred to as Myofascial Trigger Points (MTrPs) and can be classified by two states: active MTrPs (A-MTrPs) and latent MtrPs (L-MTrPs). Treatment for MPS involves massage therapy, acupuncture, and injections or painkillers. Given the subjectivity of patient pain quantification, MPS can often lead to mistreatment or drug misuse. A deterministic way …

Trojan Detection Expansion Of Structural Checking, Zachary Chapman

Graduate Theses and Dissertations

With the growth of the integrated circuit (IC) market, there has also been a rise in demand for third-party soft intellectual properties (IPs). However, the growing use of such Ips makes it easier for adversaries to hide malicious code, like hardware Trojans, into these designs. Unlike software Trojan detection, hardware Trojan detection is still an active research area. One proposed approach to this problem is the Structural Checking tool, which can detect hardware Trojans using two methodologies. The first method is a matching process, which takes an unknown design and attempts to determine if it might contain a Trojan by …

Decentralized Machine Learning On Blockchain: Developing A Federated Learning Based System, Nikhil Sridhar

Master's Theses

Traditional Machine Learning (ML) methods usually rely on a central server to per-
form ML tasks. However, these methods have problems like security risks, data
storage issues, and high computational demands. Federated Learning (FL), on the
other hand, spreads out the ML process. It trains models on local devices and then
combines them centrally. While FL improves computing and customization, it still
faces the same challenges as centralized ML in security and data storage.

This thesis introduces a new approach combining Federated Learning and Decen-
tralized Machine Learning (DML), which operates on an Ethereum Virtual Machine
(EVM) compatible blockchain. The …

Qasm-To-Hls: A Framework For Accelerating Quantum Circuit Emulation On High-Performance Reconfigurable Computers, Anshul Maurya

Theses and Dissertations

High-performance reconfigurable computers (HPRCs) make use of Field-Programmable Gate Arrays (FPGAs) for efficient emulation of quantum algorithms. Generally, algorithm-specific architectures are implemented on the FPGAs and there is very little flexibility. Moreover, mapping a quantum algorithm onto its equivalent FPGA emulation architecture is challenging. In this work, we present an automation framework for converting quantum circuits to their equivalent FPGA emulation architectures. The framework processes quantum circuits represented in Quantum Assembly Language (QASM) and derives high-level descriptions of the hardware emulation architectures for High-Level Synthesis (HLS) on HPRCs. The framework generates the code for a heterogeneous architecture consisting of a …

Automated Medical Notes Labelling And Classification Using Machine Learning, Akhil Prabhakar Thota

Electronic Theses, Projects, and Dissertations

The amount of data generated in medical records, especially in a modern context, is growing significantly. As the amount of data grows, it is very useful to classify the data into relevant classes for further interventions. Different methods that are not automated are very time-consuming and require manual effort have been tried for this before.

Recently deep learning has been used for this task but due to the complexity of the dataset, specifically due to inter-class similarities in the dataset and specific terminology having different meanings in medical contexts has caused significant problems in having a definitive approach to medical …

Melanoma Detection Based On Deep Learning Networks, Sanjay Devaraneni

Electronic Theses, Projects, and Dissertations

Our main objective is to develop a method for identifying melanoma enabling accurate assessments of patient’s health. Skin cancer, such as melanoma can be extremely dangerous if not detected and treated early. Detecting skin cancer accurately and promptly can greatly increase the chances of survival. To achieve this, it is important to develop a computer-aided diagnostic support system. In this study a research team introduces a sophisticated transfer learning model that utilizes Resnet50 to classify melanoma. Transfer learning is a machine learning technique that takes advantage of trained models, for similar tasks resulting in time saving and enhanced accuracy by …

Classification Of Thorax Diseases From Chest X-Ray Images, Sharad Jayusukhbhai Dobariya

Electronic Theses, Projects, and Dissertations

Chest X-ray images are crucial for medical decisions and patient care. However, their manual interpretation is time-consuming and prone to human error. This project aims to create an automated system that uses deep learning techniques to classify thorax disease from chest X-ray images. We are using the NIH Chest X-Ray Dataset, which contains many annotated images, as input data for this project. This approach uses UNet architecture as its classification layer. UNet architecture is well-known for its efficiency in image segmentation. Adding residual blocks enhances this approach's ability to classify images. The goal of this project is to create a …

A Modular Framework For Surface-Embedded Actuation And Optical Sensing In Soft Robots., Paul Bupe Jr

Electronic Theses and Dissertations

This dissertation explores the development and integration of modular technologies in soft robotics, with a focus on the OptiGap sensor system. OptiGap serves as a simple, flexible, cost-effective solution for real-time sensing of bending and deformation, validated through simulation and experimentation. Working as part of an emerging category of soft robotics called Soft, Curved, Reconfigurable, Anisotropic Mechanisms, or SCRAMs, this research also introduces the Thermally-Activated SCRAM Limb (TASL) technology, which employs shape-memory alloy (SMA) wire embedded in curved sheets for surface actuation and served as the initial inspiration for OptiGap. In addition, the EneGate system is presented as a complementary …

Hypothyroid Disease Analysis By Using Machine Learning, Sanjana Seelam

Electronic Theses, Projects, and Dissertations

Thyroid illness frequently manifests as hypothyroidism. It is evident that people with hypothyroidism are primarily female. Because the majority of people are unaware of the illness, it is quickly becoming more serious. It is crucial to catch it early on so that medical professionals can treat it more effectively and prevent it from getting worse. Machine learning illness prediction is a challenging task. Disease prediction is aided greatly by machine learning. Once more, unique feature selection strategies have made the process of disease assumption and prediction easier. To properly monitor and cure this illness, accurate detection is essential. In order …

Classification Of Large Scale Fish Dataset By Deep Neural Networks, Priyanka Adapa

Electronic Theses, Projects, and Dissertations

The development of robust and efficient fish classification systems has become essential to preventing the rapid depletion of aquatic resources and building conservation strategies. A deep learning approach is proposed here for the automated classification of fish species from underwater images. The proposed methodology leverages state-of-the-art deep neural networks by applying the compact convolutional transformer (CCT) architecture, which is famous for faster training and lower computational cost. In CCT, data augmentation techniques are employed to enhance the variability of the training data, reducing overfitting and improving generalization. The preliminary outcomes of our proposed method demonstrate a promising accuracy level of …

Lung Lesion Segmentation Using Deep Learning Approaches, Sree Snigdha Tummala

Electronic Theses, Projects, and Dissertations

The amount of data generated in the medical imaging field, especially in a modern context, is growing significantly. As the amount of data grows, it's prudent to make use of automated techniques that can leverage datasets to solve problems that are error-prone or have inconsistent solutions.

Deep learning approaches have gained traction in medical imaging tasks due to their superior performance with larger datasets and ability to discern the intricate features of 3D volumes, a task inefficient if done manually. Specifically for the task of lung nodule segmentation, several different methods have been tried before such as region growing etc. …

Machine Learning For Kalman Filter Tuning Prediction In Gps/Ins Trajectory Estimation, Peter Wright

Electronic Theses, Projects, and Dissertations

This project is an exploration and implementation of an application using Machine Learning (ML) and Artificial Intelligence (AI) techniques which would be capable of automatically tuning Kalman-Filter parameters used in post-flight trajectory estimation software at Edwards Air Force Base (EAFB), CA. The scope of the work in this paper is to design and develop a skeleton application with modular design, where various AI/ML modules could be developed to plug-in to the application for tuning-switch prediction.

A New Algorithm For Encounter Generation: Encounters From Actual Trajectories (Enact), James Anthony Ritchie Iii

Theses and Dissertations

There is ongoing research at the Federal Aviation Administration (FAA) and other private industries to examine a concept for delegated separation in multiple classes of airspace to allow unmanned aircraft systems (UAS) to remain well clear of other aircraft. Detect and Avoid (DAA) capabilities are one potential technology being examined to maintain separation. To evaluate these DAA capabilities, input traffic scenarios are needed, but current approaches are limited by the breadth of the traffic recordings available. This thesis derives a new mathematical algorithm that uses great circle navigation equations in an Earth spherical model and an accurate aircraft performance model …

Screensafefuture: A Parent-Empathetic And Practical Mhealth Application For Toddlers' Brain Development Addressing Screen-Addiction Challenges, Nafisa Anjum

Master of Science in Information Technology Theses

The surging incidents of infants and toddlers screen addiction in the United States are becoming a pressing concern due to its detrimental and compound impact on cognitive development, mental health, and physical growth. To address this era's critical child health and human development problem, we propose an innovative mHealth application-- ScreenSafeFuture-- in this paper. ScreenSafeFuture provides practical and parent-friendly solutions that seamlessly fit into parents' busy lifestyles, also acknowledging the effectiveness and convenience of smartphones as a healthcare tool. Our offering includes essential features designed to enhance the experience between parents and their children under 3 years old. With an …

Deep Learning-Based Automatic Stereology For High- And Low-Magnification Images, Hunter Morera

USF Tampa Graduate Theses and Dissertations

Quantification of the true number of stained cells in specific brain regions is an important metric in many fields of biomedical research involving cell degeneration, cytotoxicology, cellular inflammation, and drug development for a wide range of neurological disorders and mental illnesses. Unbiased stereology is the current state-of-the-art method for collecting the cell count data from tissue sections. These studies require trained experts to manually focus through a z-stack of microscopy images and count (click) on a hundred or more cells per case, making this approach time consuming (~1 hour per case) and prone to human error (i.e., inter-rater variability). Thus, …

Analyzing Multi-Robot Leader-Follower Formations In Obstacle-Laden Environments, Zachary J. Hinnen

USF Tampa Graduate Theses and Dissertations

Observations in biological formation from nature likes flocks of birds, herds of mammals and packs of wolves have inspired the innovation of robotic architectures. This thesis presents an approach that aims to use robotic systems to mimic leader-follower behaviors in the navigation and formation of sparse and dense environments. The goal of this work is to extend and further analyze the original work of Weitzenfeld et al [3] to evaluate new swarm and pack based multi-robot architectures with the inclusion of obstacle avoidance and variations in group formations. The multiple robot architecture is based off a wolf pack with a …

Cyber-Physical Multi-Robot Systems In A Smart Factory: A Networked Ai Agents Approach, Zixiang Nie

USF Tampa Graduate Theses and Dissertations

This dissertation focuses on addressing the technical challenges of non-stationarity in smart factories through the use of cyber-physical AI agents. Industry 4.0 and smart manufacturing with smart factories as a central role, have a growing demand for Just-in-Time (JIT) and on-demand production, as well as mass customization—all while maintaining high productivity, resource efficiency and resilience. This research positions Multi-Robot Systems (MRS)-driven smart factories. The heterogeneous production and transportation robots in an MRS collaborate to form multiple real-time adjusted production flows achieving the flexibility to accommodate such on-demand, mass customization.

However, the implementation of MRS introduces new sets of challenges, including …

Enabling Intelligent Network Management Through Multi-Agent Systems: An Implementation Of Autonomous Network System, Petro Mushidi Tshakwanda

Electrical and Computer Engineering ETDs

This Ph.D. dissertation presents a pioneering Multi-Agent System (MAS) approach for intelligent network management, particularly suited for next-generation networks like 5G and 6G. The thesis is segmented into four critical parts. Firstly, it contrasts the benefits of agent-based design over traditional micro-service architectures. Secondly, it elaborates on the implementation of network service agents in Python Agent Development Environment (PADE), employing machine learning and deep learning algorithms for performance evaluation. Thirdly, a new scalable approach, Scalable and Efficient DevOps (SE-DO), is introduced to optimize agent performance in resource-constrained settings. Fourthly, the dissertation delves into Quality of Service (QoS) and Radio Resource …