Computer Sciences Commons^™

Development And Application Of Self-Supervised Machine Learning For Smoke Plume And Active Fire Identification From The Fire Influence On Regional To Global Environments And Air Quality Datasets, Nicholas Lahaye, Anastasija Easley, Kyongsik Yun, Hugo Lee, Erik Linstead, Michael J. Garay, Olga V. Kalashnikova

Engineering Faculty Articles and Research

Fire Influence on Regional to Global Environments and Air Quality (FIREX-AQ) was a field campaign aimed at better understanding the impact of wildfires and agricultural fires on air quality and climate. The FIREX-AQ campaign took place in August 2019 and involved two aircraft and multiple coordinated satellite observations. This study applied and evaluated a self-supervised machine learning (ML) method for the active fire and smoke plume identification and tracking in the satellite and sub-orbital remote sensing datasets collected during the campaign. Our unique methodology combines remote sensing observations with different spatial and spectral resolutions. With as much as a 10% …

From Image Enhancement To Model Protection Integrating Generative Ai And Secure Learning In Computer Vision, Mohammad Shahab Uddin

Electrical & Computer Engineering Theses & Dissertations

This dissertation aims to address critical challenges in the field of computer vision and machine learning, focusing on three key areas: image translation, denoising, and model security. The research encompasses novel methodologies and models that significantly advance existing techniques. This dissertation will not only provide valuable contributions to the academic community but also hold significant potential for practical applications in domains ranging from surveillance to autonomous systems.

Consequently, this dissertation proposes three goals. First, we present new approaches for converting optical videos to infrared videos using deep learning. To apply powerful deep learning based algorithms for object detection and classification …

Standardizing Canine Breed Data In Veterinary Records Is Challenging, But Computer Vision Offers An Alternative Perspective On Breed Assignment, Glenvelis Perez, Yixuan He, Zihan Lyu, Yilin Chen, Nicholas Howe, Halie M. Rando

Computer Science: Faculty Publications

Dog breed is fundamental health information, especially in the context of breed-linked diseases. The standard-ization of breed terminology across health records is necessary to leverage the big data revolution for veterinary research. Breed can also inform clinical decision making. However, client-reported breeds vary in their reliability depending on how breed was determined. Surprisingly, research in computer science reports that AI can assign breed to dogs with over 90% accuracy from a photograph. Here, we explore the extent to which current research in AI is relevant to breed assignment or validation in veterinary contexts. This review provides a primer on approaches …

Accelerated Multiobjective Calibration Of Fused Deposition Modeling 3d Printers Using Multitask Bayesian Optimization And Computer Vision, Craig S. Ganitano, Benji Maruyama, Gilbert L. Peterson

Faculty Publications

Proper process parameter calibration is critical to the success of fused deposition modeling (FDM) three-dimensional (3D) printing, but is time-consuming and requires expertise. While existing systems for autonomous calibration have demonstrated success in calibrating for a single objective, users may need to balance multiple conflicting objectives. Herein, an easily deployable, camera-based system for autonomous calibration of FDM printers that optimizes for both part quality and completion time is presented. Autonomous calibration is achieved through a novel, multifaceted computer vision characterization and a multitask learning extension to Bayesian optimization. The system is demonstrated on four popular filament types using two distinct …

Scalable, Secure, And Adaptable Perception Systems Through Adversarial Analysis And Federated Fine-Tuning, Arkajyoti Mitra

Computer Science and Engineering Dissertations - Archive

Perception systems are fundamental to intelligent machines, enabling them to sense, understand, and interpret complex environments. However, as perception increasingly underpins critical applications such as autonomous vehicles, IoT healthcare devices, and smart trading platforms, challenges related to security, scalability, and environmental understanding have become more pressing. This work addresses three core research questions: (1) How can we identify, analyze, and mitigate adversarial vulnerabilities in perception systems to ensure reliable operation under adversarial conditions? (2.1) How can AVPS models be efficiently scaled and fine-tuned across decentralized and resource-constrained environments while preserving privacy and performance? (2.2) How can we scale generative models …

Comparative Evaluation Of Traditional Machine Learning And Deep Cnn Models For Static Hand Gesture Recognition, Anamol Khadka, Prit Desai

Computer Science and Engineering Student Research - Archive

Hand gesture recognition plays a vital role in facilitating natural and intuitive human-computer interaction, with applications ranging from sign language translation to touchless control systems. This study presents a comparative evaluation of traditional machine learning models and a deep convolutional neural network (CNN) for static hand gesture classification. The experimental dataset comprises 24,000 training images and 6,000 testing images, spanning 20 gesture classes. Traditional models, including k-Nearest Neighbors (KNN) and Support Vector Machines (SVM), utilize handcrafted features such as convex hull, convexity defects, and Hu moments. In contrast, the deep learning approach fine-tunes a ResNet18 architecture to learn features directly …

Human Perception Of Ai Capabilities At Classifying Perturbed Roadway Signs, Katherine R. Garcia, Jing Chen, Yanru Xiao, Scott Mishler, Cong Wang, Bin Hu

Psychology Faculty Publications

Artificial Intelligence (AI) is crucial to numerous functions required for driving automation systems, including the computer vision techniques used to detect the roadway environment and make real-time decisions. However, the images used as inputs to the AI system may be maliciously perturbed, or manipulated, causing the AI system to make an incorrect classification. In this study, we examined humans’ perception of the AI’s computer vision capability of classifying various road sign images, including the original images, images with two different types of malicious attacks, and images that are scrambled randomly at the pixel level. Our results showed that participants rated …

Information-Theoretic Methods For Efficient Training And Robust Evaluations In Self-Supervised Learning, Oscar Skean

Theses and Dissertations--Computer Science

Self-supervised learning (SSL) has become a cornerstone of modern machine learning, offering a scalable alternative to costly human annotation by constructing pretext tasks directly from raw data. While SSL has delivered strong results across vision, language, and multimodal domains, two major limitations persist: (1) SSL methods are often significantly slower to train than supervised counterparts, and (2) evaluation protocols remain narrow, with most studies relying on linear probing accuracy on the pretraining dataset. . These challenges are particularly acute for large language models (LLMs), where training costs and interpretability of intermediate representations are critical concerns.

In this work, we propose …

Embodied Ai For Challenging Rearrangement Tasks In The Context Of Service And Assistive Robots, Mariia Khan

Theses: Doctorates and Masters

Embodied AI explores intelligent agents that learn through interaction with their environment, aiming to replicate human-like learning processes. Achieving this requires agents capable of understanding a scene via various sensors, reasoning about their actions, and reacting accordingly. These abilities are necessary for service domestic robots to assist humans in their day-to-day activities. Embodied AI tasks can include but are not limited to: visual exploration, visual navigation, instruction following and embodied question answering, which typically consider static (unchanging) environments, where objects do not move over time. This thesis addresses one of the most challenging Embodied AI tasks – visual room rearrangement, …

Multi-Modal Graph Learning For Vision Language Model In General And Medical Domains, Xinyue Hu

Computer Science and Engineering Dissertations - Archive

Multi-modal learning has gained significant attention in deep learning for its ability to integrate and process information from multiple modalities, such as text, images, and videos. By leveraging complementary information from different modalities, it enables a more comprehensive understanding of complex data in various tasks. Simultaneously, graph learning, a prominent paradigm that models structured data as graphs, captures both local and global dependencies, providing a natural framework to represent intricate interactions and contextual relationships. When combined with multi-modal learning, these graph-based approaches have the potential to enhance feature representation and reasoning by effectively fusing heterogeneous data, leading to more robust …

Automated Methods For Estimating Blood Alcohol Concentration Level From Facial Cues, Ensiyeh Keshtkaran

Theses: Doctorates and Masters

This thesis investigates different approaches for detecting alcohol intoxication in drivers by analysing facial video data. Tackling this issue necessitates the creation of a novel dataset to overcome the limitations of existing datasets. The dataset constructed in this study is the first to include RGB video recordings of individual faces at varying levels of alcohol intoxication during simulated driving, featuring 60 participants with BAC levels ranging from 0 to 0.165 g/100ml. The constructed dataset not only supports this thesis, but also offers the broader scientific community a valuable resource for further study and development.

Building on this, this thesis presents …

Toward Embodied Navigation Through Vision And Language, Muraleekrishna Gopinathan

Theses: Doctorates and Masters

Embodied AI is a challenging but exciting field in which a robot learns to interact with human-living spaces to perform various tasks. This thesis studies the embodied navigation problem in which a robotic agent navigates in a previously unseen indoor environment based on a challenging task. In particular, the Vision-and-Language Navigation (VLN) task requires a robot to navigate based on a descriptive human-language instruction. This thesis aims to improve VLN agents on four key aspects - their understanding of the environment, training via additional data, correcting navigational errors, and predicting the layout of the environment for better planning.

First, we …

Using Quanser Platform To Introduce Engineering Technology Students To Autonomous Vehicles, Otilia Popescu, Logan Beaver, Murat Kuzlu, Krishnanand Kaipa

Engineering Technology Faculty Publications

The area of autonomous vehicles is not new, but the latest advances in various technologies gave it a new boost in the last decade and it keeps growing in interest. However, undergraduate curricula rarely include courses specific to this area, which is considered mostly an interdisciplinary graduate field. While various programs introduce students to the background needed to understand and approach the field, specific work on autonomous vehicle projects is left for extra curriculum activities or student clubs, and eventually for senior (capstone) projects. This paper presents the work of a team of electrical engineering technology students on an autonomous …

Leveraging Synthetic Data For Efficient Training Of Ai Models For Real-World Object Detection, Reinaldo A. Moraga

Graduate Research Theses & Dissertations

Modern computer vision (CV) systems largely depend on real-world data for training, which is costly in terms of time, materials, and resources. As industries push toward automation and Artificial Intelligence (AI) -driven solutions, the need for enabling more efficient model training is growing. The primary aim of this work is to explore a framework tailored for industrial applications that uses synthetic images generated from 3D models to train a CV model capable of real-world object detection. This approach seeks to reduce the time, cost, and resources typically required for training AI models with real-world data. This work presents a method …

Facial Chick Sexing: An Automated Chick Sexing System From Chick Facial Image, Marta Veganzones Rodriguez, Thinh Phan, Arthur F.A. Fernandes, Vivian Breen, Jesus Arango, Michael T. Kidd, Ngan Le

Computer Science and Computer Engineering Faculty Publications and Presentations

Chick sexing, the process of determining the gender of day-old chicks, is a critical task in the poultry industry due to the distinct roles that each gender plays in production. While effective traditional methods achieve high accuracy, color, and wing feather sexing is exclusive to specific breeds, and vent sexing is invasive and requires trained experts. To address these challenges, we propose a novel approach inspired by facial gender classification techniques in humans: facial chick sexing. This new method does not require expert knowledge and aims to reduce training time while enhancing animal welfare by minimizing chick manipulation. We develop …

Human Perception Of Ai Capabilities At Classifying Perturbed Roadway Signs, Katherine R. Garcia, Jing Chen, Yanru Xiao, Scott Mishler, Cong Wang, Bin Hu

Computer Science Faculty Publications

Artificial Intelligence (AI) is crucial to numerous functions required for driving automation systems, including the computer vision techniques used to detect the roadway environment and make real-time decisions. However, the images used as inputs to the AI system may be maliciously perturbed, or manipulated, causing the AI system to make an incorrect classification. In this study, we examined humans’ perception of the AI’s computer vision capability of classifying various road sign images, including the original images, images with two different types of malicious attacks, and images that are scrambled randomly at the pixel level. Our results showed that participants rated …

Optimizing Vgg16 Deep Learning Model With Enhanced Hunger Games Search For Logo Classification, Mohammed Hussain, Thaer Thaher, Mohamed Basel Almourad, Majdi Mafarja

All Works

Accurate classification of logos is a challenging task in image recognition due to variations in logo size, orientation, and background complexity. Deep learning models, such as VGG16, have demonstrated promising results in handling such tasks. However, their performance is highly dependent on optimal hyperparameter settings, whose fine-tuning is both labor-intensive and time-consuming. Swarm intelligence algorithms have been widely adopted to solve many highly nonlinear, multimodal problems and have succeeded significantly. The Hunger Games Search (HGS) is a recent swarm intelligence algorithm that has shown good performance across various applications. However, the standard HGS still faces limitations, such as restricted population …

The Computational Eye. Deconstructing Style In Digital Art History, Paul Guhennec, Ellen Charlesworth

Artl@s Bulletin

With the aim of grounding digital methods in the art historic tradition, this paper uses the discussions around style as a springboard to ask how digital art history can extend beyond providing quantitative confirmation of known trends to enrich our current understanding of visual cultures. Drawing from the examples throughout this issue, we explore how an analysis of computational ways of seeing—or the ‘computational eye’—can expose the underlying preoccupations and priorities of our own research.

Afin de mieux ancrer les méthodes numériques dans la tradition de l’histoire de l’art, cet article se sert des discussions récentes autour du concept de …

Decoding Neural Networks: An Information-Theoretic Guide To Interpretability, Error Analysis And Efficiency, Mackenzie J. Meni

Theses and Dissertations

This dissertation addresses critical challenges in neural network design by leveraging entropy-based techniques to improve model efficiency, interpretability, and bias reduction. Focusing on the unique demands of computer vision applications, particularly object detection and classification for real-time systems, this work introduces a series of innovative methods centered on information theory. At the core of these methods is the Probabilistic Explanations of Entropic Knowledge (PEEK) framework, a tool developed to analyze and visualize entropy distributions across feature maps. PEEK offers insights into information flow within neural networks, making it possible to pinpoint layers that contribute meaningfully to decision-making or identify those …

Collectively Advancing Deep Learning For Animal Detection In Drone Imagery: Successes, Challenges, And Research Gaps, Daniel Axford, Ferdous Sohel, Mathew A. Vanderklift, Amanda J. Hodgson

Research outputs 2022 to 2026

Drones have emerged as a powerful tool in animal detection, significantly advancing wildlife monitoring, conservation, and management by capturing high-resolution, real-time imagery over areas often inaccessible or challenging for human observers to reach. However, manual analysis of drone imagery for animal detection is labour-intensive and time-consuming. The application of deep learning methods, particularly convolutional neural networks, in automating animal detection from drone imagery has the potential to revolutionise wildlife monitoring, conservation, and management protocols. This review provides a comprehensive overview of the increasing use and prospects of deep learning in animal detection using drone imagery. It explores successful applications of …

Fast And High-Resolution View Synthesis From A Single Input Panorama, Nam Nguyen, Angela V. Chen, Theresa Zhu, Seth Johnson, Pranav Dumpa, Benjamin Geil

College of Engineering Summer Undergraduate Research Program

We introduce a novel method to convert a single input panorama into a 3D colored mesh representation of the scene. Unlike recent methods based on neural rendering, which are limited to low-resolution inputs and offline rendering, our approach supports 4k resolution inputs and real time rendering in a virtual reality headset. We first estimate a depth map and produce an initial layered depth image (LDI) representation. We fill unseen regions behind objects by iteratively cutting and inpainting the LDI. We then convert the LDI into an optimized, texture mapped mesh to achieve a compact representation

Improving Out-Of-Distribution Detection With Disentangled Foreground And Background Features, Choubo Ding, Guansong Pang

Research Collection School Of Computing and Information Systems

Detecting out-of-distribution (OOD) inputs is a principal task for ensuring the safety of deploying deep-neural-network classifiers in open-set scenarios. OOD samples can be drawn from arbitrary distributions and exhibit deviations from in-distribution (ID) data in various dimensions, such as foreground features (e.g., objects in CIFAR100 images vs. those in CIFAR10 images) and background features (e.g., textural images vs. objects in CIFAR10). Existing methods can confound foreground and background features in training, failing to utilize the background features for OOD detection. This paper considers the importance of feature disentanglement in out-of-distribution detection and proposes the simultaneous exploitation of both foreground and …

Etdsuite: A Toolkit To Mine Electronic Theses And Dissertations To Enrich Scholarly Big Data Using Natural Language Processing And Computer Vision, Muntabir Hasan Choudhury

Computer Science Theses & Dissertations

In the past decades, there has been a growing interest in mining scientific documents to obtain domain knowledge automatically. One of the understudied types of scientific documents is Electronic Theses and Dissertations (ETDs), as ETDs have distinct features compared with conference proceedings and journal articles. ETDs usually serve as partial requirements of academic degrees for students pursuing higher education. They are book-length documents (i.e., 100 – 400 pages long), and the topics may shift across chapters, exhibit the significant contribution of a student’s research over the entire degree pursuing period, and have unique metadata schema and page layouts. However, the …

Dirt/Μ: Automated Extraction Of Root Hair Traits Using Combinatorial Optimization, Peter Pietrzyk, Neen Phan-Udom, Chartinun Chutoe, Lise Pingault, Ankita Roy, Marc Libault, Patompong Johns Saengwilai, Alexander Bucksch

Department of Entomology: Faculty Publications

As with phenotyping of any microscopic appendages, such as cilia or antennae, phenotyping of root hairs has been a challenge due to their complex intersecting arrangements in two-dimensional images and the technical limitations of automated measurements. Digital Imaging of Root Traits at Microscale (DIRT/μ) is a newly developed algorithm that addresses this issue by computationally resolving intersections and extracting individual root hairs from two-dimensional microscopy images. This solution enables automatic and precise trait measurements of individual root hairs. DIRT/μ rigorously defines a set of rules to resolve intersecting root hairs and minimizes a newly designed cost function to combinatorically identify …

Two-Dimensional Quantum Material Identification Via Self-Attention And Soft-Labeling In Deep Learning, Xuan Bac Nguyen, Apoorva Bisht, Benjamin Thompson, Hugh O.H. Churchill, Khoa Luu, Samee U. Khan

Electrical Engineering and Computer Science Faculty Publications and Presentations

Detecting two-dimensional (2D) materials in silicon chips presents a significant challenge in the field of quantum machines due to the difficulty of data collection. Specifically, among thousands of flakes, not all flakes are useful or well-annotated, resulting in noisy and hard samples within the dataset, which challenges the deep neural network (DNN) to learn. To address this problem, we propose a novel method for identifying quantum 2D flakes even when there is a high rate of missing annotations in the input images. In particular, we first propose a new mechanism for automatically detecting false negative flakes that are missing annotations. …

Robotic Odor Source Localization Using Vision And Olfaction Sensing, Sunzid Hassan

Master's Theses

Robotic Odor Source Localization (ROSL) technology allows autonomous agents like robots to find an odor source in unknown environments. A successful odor source location depends crucially on an effective navigation algorithm that directs the robot towards the odor source. This thesis is a combination of three projects. First, we detail development of a versatile multi-modal robotic platform for ROSL real-world ROSL experimentation and discussed real-world validation of a traditional olfactionbased ROSL algorithm. Secondly, we introduced vision in ROSL by proposing a fusion navigation algorithm that integrates deep-learning enabled vision and olfaction-based navigation. This hybrid approach tackles challenges such as turbulent …

Predictive Filtering-Based Image Inpainting, Xiaoguang Li

Theses and Dissertations

Image inpainting is an important challenge in the computer vision field. The primary goal of image inpainting is to fill in the missing parts of an image. This technique has many real-life uses including fixing old photographs and restoring ancient artworks, e.g., the degraded Dunhuang frescoes. Moreover, image inpainting is also helpful in image editing. It has the capability to eliminate unwanted objects from images while maintaining a natural and realistic appearance, e.g., removing watermarks and subtitles. Disregarding the fact that image inpainting expects the restored result to be identical to the original clean one, existing deep generative inpainting methods …

Challenges And Practices Of Deep Learning Model Reengineering: A Case Study On Computer Vision, Wenxin Jiang, Vishnu Banna, Naveen Vivek, Abhinav Goel, Nicholas Synovic, George K. Thiruvathukal, James C. Davis

Computer Science: Faculty Publications and Other Works

Many engineering organizations are reimplementing and extending deep neural networks from the research community. We describe this process as deep learning model reengineering. Deep learning model reengineering — reusing, replicating, adapting, and enhancing state-of-the-art deep learning approaches — is challenging for reasons including under-documented reference models, changing requirements, and the cost of implementation and testing.

Scene Text Detection And Recognition Via Discriminative Representation, Liang Zhao

Theses and Dissertations

Scene texts refer to arbitrary text presented in an image captured by a camera in the real world. The tasks of scene text detection and recognition from complex images play a crucial role in computer vision, with potential applications in scene understanding, information retrieval, robotics, autonomous driving, etc. Despite the notable progress made by existing deep-learning methods, achieving accurate text detection and recognition remains challenging for robust real-world applications. The challenges in scene text detection and recognition stem from: 1) diverse text shapes, fonts, colors, styles, layouts, etc.; 2) countless combinations of characters with unfixed attributes for complete detection, coupled …

Development Of Feature Extraction Models To Improve Image Analysis Applications In Cancer, Yu Shi

Dissertations and Doctoral Documents, University of Nebraska-Lincoln, 2023–

Cancer poses a significant global health challenge. With an estimated 20 million new cases diagnosed worldwide in 2022 and 9.7 million fatalities attributable to the disease, the economic burden of cancer is immense. It impacts healthcare systems and imposes substantial costs for its care on patients and their families. Despite advancements in early detection, prevention, and treatment that have reduced overall cancer mortality rates, the growing prevalence of cancer, particularly among younger individuals, remains a pressing issue.

Recent advancements in medical imaging technology have progressed significantly with the help of emerging computer vision and artificial intelligence (AI) technology. Despite these …