Deep Face Morph Detection Based On Wavelet Decomposition, Poorya Aghdaie

Graduate Theses, Dissertations, and Problem Reports

Morphed face images are maliciously used by criminals to circumvent the official process for receiving a passport where a look-alike accomplice embarks on requesting a passport. Morphed images are either synthesized by alpha-blending or generative networks such as Generative Adversarial Networks (GAN). Detecting morphed images is one of the fundamental problems associated with border control scenarios. Deep Neural Networks (DNN) have emerged as a promising solution for a myriad of applications such as face recognition, face verification, fake image detection, and so forth. The Biometrics communities have leveraged DNN to tackle fundamental problems such as morphed face detection. In this …

Object Detection And Classification In The Visible And Infrared Spectrums, Domenick D. Poster

Graduate Theses, Dissertations, and Problem Reports

The over-arching theme of this dissertation is the development of automated detection and/or classification systems for challenging infrared scenarios. The six works presented herein can be categorized into four problem scenarios. In the first scenario, long-distance detection and classification of vehicles in thermal imagery, a custom convolutional network architecture is proposed for small thermal target detection. For the second scenario, thermal face landmark detection and thermal cross-spectral face verification, a publicly-available visible and thermal face dataset is introduced, along with benchmark results for several landmark detection and face verification algorithms. Furthermore, a novel visible-to-thermal transfer learning algorithm for face landmark …

Enhancing Vehicular Perception: A Comprehensive Analysis Of Sensor Fusion Performance Through Weighted Averages And Fuzzy C-Means For Optimal Data Association, Zachary Brian Flanigan

Graduate Theses, Dissertations, and Problem Reports

This work explores the implementation of sensor fusion and data association for autonomous vehicle design. Advancements in Adaptive Driver Assistance System (ADAS) technology have driven the development of perception algorithms required for higher levels of autonomy in vehicles. Perception algorithms process data collected from radar, camera, and LiDAR sensors to generate a complete model of the ego vehicle’s surrounding environment. Fusion of data from these sensors is important for accurate measurement of longitudinal and lateral distances to surrounding objects. Sensor fusion associates sensor detections to each other through different data association techniques. Data association techniques can consist of independent assignment …

System Analysis Of An Internal Combustion Engine (Ice) – Solid Oxide Fuel Cell (Sofc) Hybrid Cycle, Jose Javier Colon Rodriguez

Graduate Theses, Dissertations, and Problem Reports

Due to the intermittent nature of renewable energy and the rigid operation of existing coal plants, the need for flexible power generation technology is eminent. Hybrid energy systems have shown potential for flexible, grid following dynamics while maintaining higher efficiencies. The work below focuses on the performance analysis of a proposed 100 kW pressurized Internal Combustion Engine (ICE) and Solid Oxide Fuel Cell (SOFC) hybrid system. The un-utilized fuel from the SOFC stack provided the chemical energy to operate the engine. A turbocharger was used to deliver the necessary air flow for both the stack and engine. An external reformer …

Optimal Deployment Of Air Vehicle As Communication Relay For Multiple Ground Vehicles, Juan David Pabon Arias

Graduate Theses, Dissertations, and Problem Reports

Heterogeneous teams of both air and ground mobile vehicles can combine the advantages of mobility, sensing capability, and operation time when performing complex tasks. However, when ground vehicles operate in cluttered environments with randomized obstacles, they may experience line of sight (LoS) obstructions and loss of communication due to those obstacles. To mitigate this issue, an airborne relay can be positioned in the vicinity of the ground vehicles to aid communication by establishing two-hop communication links between the vehicles.

This thesis develops an analytical framework to calculate the probability of spanning a two-hop communication between a pair of ground vehicles …

Community And Key Player Detection For Disrupting Illicit Drug Supply Networks In Social Media Platforms – Especially On Instagram, Akassi Rachel Niamke Aman

Graduate Theses, Dissertations, and Problem Reports

This thesis focuses on the pressing issue of illicit drug trafficking and its impact on public health and safety at a global level. With the advent of digital technologies and social media platforms, combating drug trafficking has become increasingly challenging for law enforcement and researchers alike. Among these platforms, Instagram, a popular photo and video-sharing social networking platform, has emerged as a prominent hub for drug trafficking activities.

In this study, we delve into the effectiveness of community and key player detection algorithms in identifying and disrupting illicit drug supply networks on Instagram. To conduct our research, we collected real …

Dynamic Modeling, Data Reconciliation, Parameter Estimation, And Health Monitoring Of A Supercritical Power Plant, Katherine Grace Hedrick

Graduate Theses, Dissertations, and Problem Reports

With the introduction of a larger portion of renewable sources of power coming onto the U.S. power grid in recent decades, the operational strategy of coal-fired power plants has changed significantly to focus more on flexibility in response to the changing energy market. This has naturally led to different operational challenges. Many of these challenges are focused on the boilers within these plants, as they are producing more emissions and experiencing increased damage during load-following, which in turn leads to increased costs from penalties for not achieving emission standards or maintenance costs as boilers accumulate damage from the cycling behavior. …

Computational Mechanisms Of Face Perception, Jinge Wang

Graduate Theses, Dissertations, and Problem Reports

The intertwined history of artificial intelligence and neuroscience has significantly impacted their development, with AI arising from and evolving alongside neuroscience. The remarkable performance of deep learning has inspired neuroscientists to investigate and utilize artificial neural networks as computational models to address biological issues. Studying the brain and its operational mechanisms can greatly enhance our understanding of neural networks, which has crucial implications for developing efficient AI algorithms. Many of the advanced perceptual and cognitive skills of biological systems are now possible to achieve through artificial intelligence systems, which is transforming our knowledge of brain function. Thus, the need for …

Investigating The Impact Of Demographic Factors On Contactless Fingerprint Interoperability, Aeddon David Berti

Graduate Theses, Dissertations, and Problem Reports

Improvements in contactless fingerprinting have resulted in contactless fingerprints becoming a faster and more convenient alternative to contact fingerprints. The interoperability between contactless fingerprints and contact fingerprints and how demographic factors can change interoperability has been challenging since COVID-19; the need for hygienic alternatives has only grown because of the sudden focus during the pandemic. Past work has shown issues with the interoperability of contactless prints from kiosk devices and phone fingerprint collection apps. Demographic bias in photography for facial recognition could affect photographed fingerprints. The paper focuses on evaluating match performance between contact and contactless fingerprints and evaluating match …

Simulations Of Implementation Of Advanced Communication Technologies, Ivy Yousuf Moutushi

Graduate Theses, Dissertations, and Problem Reports

Wireless communication systems have seen significant advancements with the introduction of 3G, 4G, and 5G mobile standards. Since the simulation of entire systems is complex and may not allow evaluation of the impact of individual techniques, this thesis presents techniques and results for simulating the performance of advanced signaling techniques used in 3G, 4G, and 5G systems, including Code division multiple access (CDMA), Multiple Input Multiple Output (MIMO) systems, and Low-Density Parity Check (LDPC) codes. One implementation issue that is explored is the use of quantized Analog to Digital Converter (ADC) outputs and their impact on system performance.

Code division …

Component Optimization Of A Parallel P4 Hybrid Electric Vehicle Utilizing An Equivalent Consumption Minimization Strategy, Holden Ryan Fraser

Graduate Theses, Dissertations, and Problem Reports

Advancements in battery and electric motor technology have driven the development of hybrid electric vehicles to improve fuel economy. Hybrid electric vehicles can utilize an internal combustion engine and an electric motor in many configurations, requiring the development of advanced energy management strategies for a range of component configurations. The Equivalent Consumption Minimization Strategy (ECMS) is an advanced energy management strategy that can be calculated in-vehicle in real-time operation. This energy management strategy uses an equivalence factor to equate electrical to mechanical power when performing the torque split determination between the internal combustion engine and electric motor. This equivalence factor …

Framework For Data Acquisition And Fusion Of Camera And Radar For Autonomous Vehicle Systems, Clay Edward Vincent

Graduate Theses, Dissertations, and Problem Reports

The primary contribution is the development of the data collection testing methodology for autonomous driving systems of a hybrid electric passenger vehicle. As automotive manufacturers begin to develop adaptive cruise control technology in vehicles, progress is being made toward the development of fully-autonomous vehicles. Adaptive cruise control capability is classified into five levels defined by the Society of Automotive Engineering. Some vehicles under development have attained higher levels of autonomy, but the focus of most commercial development is Level 2 autonomy. As the level of autonomy increases, the sensor technology becomes more advanced with a sensor suite which includes radar, …

Imitation Learning For Swarm Control Using Variational Inference, Hafeez Olafisayo Jimoh

Graduate Theses, Dissertations, and Problem Reports

Swarms are groups of robots that can coordinate, cooperate, and communicate to achieve tasks that may be impossible for a single robot. These systems exhibit complex dynamical behavior, similar to those observed in physics, neuroscience, finance, biology, social and communication networks, etc. For instance, in Biology, schools of fish, swarm of bacteria, colony of termites exhibit flocking behavior to achieve simple and complex tasks. Modeling the dynamics of flocking in animals is challenging as we usually do not have full knowledge of the dynamics of the system and how individual agent interact. The environment of swarms is also very noisy …

Machine Learning For Biosensors, Gayathri Anapanani

Graduate Theses, Dissertations, and Problem Reports

Biosensors have become increasingly popular as diagnostic tools due to their ability to detect and quantify biological analytes in a wide range of applications. With the growing demand for faster and more reliable biosensing devices, machine learning has become a valuable tool in enhancing biosensor performance. In this report, we review recent progress in the application of machine learning to biosensors. We discuss the potential benefits of using machine learning in biosensors, including improved sensitivity, selectivity, and accuracy. We also discuss the various machine learning techniques that have been applied to biosensors, including data preprocessing, feature extraction, and classification and …

Generative Adversarial Network And Its Application In Aerial Vehicle Detection And Biometric Identification System, Moktari Mostofa

Graduate Theses, Dissertations, and Problem Reports

In recent years, generative adversarial networks (GANs) have shown great potential in advancing the state-of-the-art in many areas of computer vision, most notably in image synthesis and manipulation tasks. GAN is a generative model which simultaneously trains a generator and a discriminator in an adversarial manner to produce real-looking synthetic data by capturing the underlying data distribution. Due to its powerful ability to generate high-quality and visually pleasing
results, we apply it to super-resolution and image-to-image translation techniques to address vehicle detection in low-resolution aerial images and cross-spectral cross-resolution iris recognition. First, we develop a Multi-scale GAN (MsGAN) with multiple …

Ai-Driven Security Constrained Unit Commitment Using Predictive Modeling And Eigen Decomposition, Talha Iqbal

Graduate Theses, Dissertations, and Problem Reports

Security Constrained Unit Commitment (SC-UC) is a complex large scale mix integer constrained optimization problem solved by Independent System Operators (ISOs) in the daily planning of the electricity markets. After receiving offers and bids, ISOs have only few hours to clear the day-ahead electricity market. It requires a lot of computational effort and a reasonable time to solve a large-scale SC-UC problem. However, exploiting the fact that a UC problem is solved several times a day with only minor changes in the system data, the computational effort can be reduced by learning from the historical data and identifying the patterns …