Engineering Commons^™

Machine Learning With Topological Data Analysis, Ephraim Robert Love

Doctoral Dissertations

Topological Data Analysis (TDA) is a relatively new focus in the fields of statistics and machine learning. Methods of exploiting the geometry of data, such as clustering, have proven theoretically and empirically invaluable. TDA provides a general framework within which to study topological invariants (shapes) of data, which are more robust to noise and can recover information on higher dimensional features than immediately apparent in the data. A common tool for conducting TDA is persistence homology, which measures the significance of these invariants. Persistence homology has prominent realizations in methods of data visualization, statistics and machine learning. Extending ML with …

Expanding Image Datasets For Deep Learning By Evaluating Independence Through Coefficient Correlation And Mean-Squared Error, Ayman Yousef

Biomedical Engineering Undergraduate Honors Theses

With deep learning being leveraged more regularly in the field of image classification, particularly in medical imaging, network optimizations have become a field in and of itself. With open source, comprehensive medical image datasets few and far, computational dataset expansion has become a useful tool for researchers without the ability to further manually collect data. However, with the rich amount of data that imaging modalities like multi-photon microscopy collect at a time, there is potential to expand datasets through proper utilization of this data that often time goes unused. Previous deep learning studies have shown that improper expansion can conflate …

Teachability And Interpretability In Reinforcement Learning, Jeevan Rajagopal

Department of Computer Science and Engineering: Dissertations, Theses, and Student Research

There have been many recent advancements in the field of reinforcement learning, starting from the Deep Q Network playing various Atari 2600 games all the way to Google Deempind's Alphastar playing competitively in the game StarCraft. However, as the field challenges more complex environments, the current methods of training models and understanding their decision making become less effective. Currently, the problem is partially dealt with by simply adding more resources, but the need for a better solution remains.

This thesis proposes a reinforcement learning framework where a teacher or entity with domain knowledge of the task to complete can assist …

Raw Depth Image Enhancement Using A Neural Network, Xuan Xie

All Graduate Theses and Dissertations, Spring 1920 to Summer 2023

The term image is often used to denote a data format that records information about a scene’s color. This dissertation object focuses on a similar format for recording distance information about a scene, “depth images”. Depth images have been used extensively in consumer-level applications, such as Apple’s Face ID, based on depth images for face recognition.

However, depth images suffer from low precision and high errors, and some post-processing techniques need to be utilized to improve their quality. Deep learning, or neural networks, are frameworks that use a series of hierarchically arranged nonlinear networks to process input data. Although each …

Artificial Intelligence-Driven Remaining Useful Life Prediction Of A Machinery-A Review, Dvij Barot, Honey Sharma, Mahika Yadav, Pooja Kamat

Library Philosophy and Practice (e-journal)

The Remaining Useful Life of a machine is very useful statistical information for the operator and manufacturer. It provides a very clear perspective to the user how long the machine can be operated and if any faults are detected how can they be prevented and ultimately increase the Remaining Useful Life. If the operators are aware of the forthcoming issues of the machine the downtime caused in the inspection, part delivery and eventually replacing parts is significantly reduced.

The paper presents a study on the remaining useful life of machinery as it is an emerging technique, starting from the year …

Zero-Bias Deep Learning Enabled Quick And Reliable Abnormality Detection In Iot, Yongxin Liu, Jian Wang, Jianqiang Li, Shuteng Niu, Houbing Song

Publications

Abnormality detection is essential to the performance of safety-critical and latency-constrained systems. However, as systems are becoming increasingly complicated with a large quantity of heterogeneous data, conventional statistical change point detection methods are becoming less effective and efficient. Although Deep Learning (DL) and Deep Neural Networks (DNNs) are increasingly employed to handle heterogeneous data, they still lack theoretic assurable performance and explainability. This paper integrates zero-bias DNN and Quickest Event Detection algorithms to provide a holistic framework for quick and reliable detection of both abnormalities and time-dependent abnormal events in Internet of Things (IoT).We first use the zero bias dense …

Deep Learning Based Sound Event Detection And Classification, Alireza Nasiri

Theses and Dissertations

Hearing sense has an important role in our daily lives. During the recent years, there has been many studies to transfer this capability to the computers. In this dissertation, we design and implement deep learning based algorithms to improve the ability of the computers in recognizing the different sound events.

In the first topic, we investigate sound event detection, which identifies the time boundaries of the sound events in addition to the type of the events. For sound event detection, we propose a new method, AudioMask, to benefit from the object-detection techniques in computer vision. In this method, we convert …

Pneumonia Radiograph Diagnosis Utilizing Deep Learning Network, Wesley O'Quinn

Honors College Theses

Pneumonia is a life-threatening respiratory disease caused by bacterial infection. The goal of this study is to develop an algorithm using Convolutional Neural Networks (CNNs) to detect visual signals for pneumonia in medical images and make a diagnosis. Although Pneumonia is prevalent, detection and diagnosis are challenging. The deep learning network AlexNet was utilized through transfer learning. A dataset consisting of 11,318 images was used for training, and a preliminary diagnosis accuracy of 72% was achieved.

Strategies In Botnet Detection And Privacy Preserving Machine Learning, Di Zhuang

USF Tampa Graduate Theses and Dissertations

Peer-to-peer (P2P) botnets have become one of the major threats in network security for serving as the infrastructure that responsible for various of cyber-crimes. Though a few existing work claimed to detect traditional botnets effectively, the problem of detecting P2P botnets involves more challenges. In this dissertation, we present two P2P botnet detection systems, PeerHunter and Enhanced PeerHunter. PeerHunter starts from a P2P hosts detection component. Then, it uses mutual contacts as the main feature to cluster bots into communities. Finally, it uses community behavior analysis to detect potential botnet communities and further identify bot candidates. Enhanced PeerHunter is an …

Bibliometric Survey On Multipurpose Face Recognition System Using Deep Learning, Priyanka Tupe-Waghmare, Rahul Raghvendra Joshi Prof.

Library Philosophy and Practice (e-journal)

Face recognition is a new concept coming up lately and flourishing in the field of network access and multimedia information systems. Since humans are at the focus of attention in variety of applications containing videos, the concept of face recognition is rising in popularity. Several areas involving network security, retrieval and indexing of the content, and compression of videos gain a lot of benefit from the face recognition systems and related technology. Controlling the access of the networks with the help of facial recognition not only makes it difficult for the hackers to steal the information but also makes the …

Towards Multiclass Damage Detection And Localization Using Limited Vibration Measurements, Sandeep Sony

Electronic Thesis and Dissertation Repository

Traditional vibration-based damage detection methods provide structural health information based on their measured data (i.e., acceleration and displacement response). Over the last few decades, various model-based and time-frequency methods have shown great promises for damage identification and localization. However, the existing methods are unable to perform satisfactorily in many situations, including the presence of limited sensor measurements and training data, detection of minor and progressive damage, and identification of multiclass damage, creating constraints to make them free of user-intervention and implemented using the modern sensors. The main objective of this thesis is to develop algorithms capable of damage identification and …

Source Localization With Machine Learning, Arjun Gupta

Electrical and Computer Engineering ETDs

Source localization with sensor arrays have found applications across domains beginning with radar and sonar, astronomy, acoustics, bio-medical devices and more recently in autonomous cars and adaptive communication systems. The knowledge of the spatial spectrum not only provide information about the source and interference but also assists in increasing signal integrity and avoid interference. This provides an added degree of freedom in the form of spatial diversity. This research investigates spatial spectrum estimation of waveforms from the signals sampled by arbitrarily distributed sensors. Conventional high resolution algorithms such as root-MuSiC fails to perform accurate source localization due to the reliance …

A Literature Survey And Bibliometric Analysis Of Application Of Artificial Intelligence Techniques On Wireless Mesh Networks, Smita R. Mahajan Mrs., Harikrishnan R Dr., Ketan Kotecha Dr.

Library Philosophy and Practice (e-journal)

Recent years have seen a surge in the use of technology for executing transactions in both online and offline modes. Various industries like banking, e-commerce, and private organizations use networks for the exchange of confidential information and resources. Network security is thus of utmost importance, with the expectation of effective and efficient analysis of the network traffic. Wireless Mesh Networks are effective in communicating information over a vast span with minimal costs. A network is evaluated based on its security, accessibility, and extent of interoperability. Artificial Intelligence techniques like machine learning and deep learning have found widespread use to solve …

A 3d Point Cloud Deep Learning Approach Using Lidar To Identify Ancient Maya Archaeological Sites, Heather Richards-Rissetto, David Newton, Aziza Al Zadjali

Department of Anthropology: Faculty Publications

Airborne light detection and ranging (LIDAR) systems allow archaeologists to capture 3D data of anthropogenic landscapes with a level of precision that permits the identification of archaeological sites in difficult to reach and inaccessible regions. These benefits have come with a deluge of LIDAR data that requires significant and costly manual labor to interpret and analyze. In order to address this challenge, researchers have explored the use of state-of-the-art automated object recognition algorithms from the field of deep learning with success. This previous research, however, has been limited to the exploration of deep learning processes that work with only 2D …

Cascaded Deep Learning Network For Postearthquake Bridge Serviceability Assessment, Youjeong Jang

Electronic Theses and Dissertations

Damages assessment of bridges is important to derive immediate response after severe events to decide serviceability. Especially, past earthquakes have proven the vulnerability of bridges with insufficient detailing. Due to lack of a national and unified post-earthquake inspection procedure for bridges, conventional damage assessments are performed by sending professional personnel to the onsite, detecting visually and measuring the damage state. To get accurate and fast damage result of bridge condition is important to save not only lives but also costs.
There have been studies using image processing techniques to assess damage of bridge column without sending individual to onsite. Convolutional …

Disaster Analysis Using Satellite Image Data With Knowledge Transfer And Semi-Supervised Learning Techniques, Palavi Jain

Reports

With the increase in frequency of disasters and crisis situations like floods, earthquake and hurricanes, the requirement to handle the situation efficiently through disaster response and humanitarian relief has increased. Disasters are mostly unpredictable in nature with respect to their impact on people and property. Moreover, the dynamic and varied nature of disasters makes it difficult to predict their impact accurately for advanced preparation of responses [104]. It is also notable that the economical loss due to natural disasters has increased in recent years, and it, along with the pure humanitarian need, is one of the reasons to research innovative …

The U-Net-Based Active Learning Framework For Enhancing Cancer Immunotherapy, Vishwanshi Joshi

Theses, Dissertations and Capstones

Breast cancer is the most common cancer in the world. According to the U.S. Breast Cancer Statistics, about 281,000 new cases of invasive breast cancer are expected to be diagnosed in 2021 (Smith et al., 2019). The death rate of breast cancer is higher than any other cancer type. Early detection and treatment of breast cancer have been challenging over the last few decades. Meanwhile, deep learning algorithms using Convolutional Neural Networks to segment images have achieved considerable success in recent years. These algorithms have continued to assist in exploring the quantitative measurement of cancer cells in the tumor microenvironment. …

Artificial Intelligence Aided Receiver Design For Wireless Communication Systems, Wenjie Xu

Theses, Dissertations and Capstones

Physical layer (PHY) design in the wireless communication field realizes gratifying achievements in the past few decades, especially in the emerging cellular communication systems starting from the first generation to the fifth generation (5G). With the gradual increase in technical requirements of large data processing and end-to-end system optimization, introducing artificial intelligence (AI) in PHY design has cautiously become a trend. A deep neural network (DNN), one of the population techniques of AI, enables the utilization of its ‘learnable’ feature to handle big data and establish a global system model. In this thesis, we exploited this characteristic of DNN as …

Improving A Network Intrusion Detection System’S Efficiency Using Model-Based Data Augmentation, Vinicius Waterkemper Lodetti

Dissertations

A network intrusion detection system (NIDS) is one important element to mitigate cybersecurity risks, the NIDS allow for detecting anomalies in a network which may be a cyberattack to a corporate network environment. A NIDS can be seen as a classification problem where the ultimate goal is to distinguish between malicious traffic among a majority of benign traffic. Researches on NIDS are often performed using outdated datasets that don’t represent the actual cyberspace. Datasets such as the CICIDS2018 address this gap by being generated from attacks and an infrastructure that reflects an up-to-date scenario.

A problem may arise when machine …

Improving Deep Learning Models In Weakly Supervised Learning Paradigms, Fariborz Taherkhani

Graduate Theses, Dissertations, and Problem Reports

Convolutional Neural Networks (CNNs) have provided promising achievements for image classification problems. However, training a CNN model relies on a large number of labeled data. Considering the vast amount of unlabeled data available on the web, it is important to make use of these data in conjunction with a small set of labeled data to train a deep learning model. In this thesis, we aim to develop some methods that can make use of unlabeled data along with partially labeled data to obtain a better generalization performance for CNN models. Moreover, it is also often easier to collect labels that …

Integration Of Deep Hashing And Channel Coding For Biometric Security And Biometric Retrieval, Veeru Talreja

Graduate Theses, Dissertations, and Problem Reports

In the last few years, the research growth in many research and commercial fields are due to the adoption of state of the art deep learning techniques. The same applies to even biometrics and biometric security. Additionally, there has been a rise in the development of deep learning techniques used for approximate nearest neighbor (ANN) search for retrieval on multi-modal datasets. These deep learning techniques knows as deep hashing (DH) integrate feature learning and hash coding into an end-to-end trainable framework. Motivated by these factors, this dissertation considers the integration of deep hashing and channel coding for biometric security and …

Next-Generation Re-Entry Aerothermodynamic Modeling Of Space Debris Using Machine Learning, Nicholas Sia

Graduate Theses, Dissertations, and Problem Reports

The number of resident space objects re-entering the atmosphere is expected to rise with increased space activity over recent years and future projections. Predicting the survival and impact location of the medium to large sized re-entering objects becomes important as they can cause on ground casualties and damage to property. Uncertainties associated with the re-entry process makes necessary a probabilistic approach, which can be computationally expensive when using high-fidelity numerical methods for estimating aerothermodynamic properties. To date, object-oriented analysis is the dominant tool used for atmospheric re-entry modeling and simulation, where aerothermodynamic coefficients are used to determine the risk a …

Multivariate Time Series Classification Of Sensor Data From An Industrial Drying Hopper: A Deep Learning Approach, Md Mushfiqur Rahman

Graduate Theses, Dissertations, and Problem Reports

In recent years, the advancement of industry 4.0 and smart manufacturing has made a large number of industrial process data attainable with the use of sensors installed in the machineries. This thesis proposes an experimental predictive maintenance framework for an industrial drying hopper so that it can detect any unusual event in the hopper which reduces the risk of erroneous fault diagnosis in the manufacturing shop floor. The experimental framework uses Deep Learning (DL) algorithms in order to classify Multivariate Time Series (MTS) data into two categories- failure or unusual events and regular events, thus formulating the problem as binary …

Deep Models For Improving The Performance And Reliability Of Person Recognition, Sobhan Soleymani

Graduate Theses, Dissertations, and Problem Reports

Deep models have provided high accuracy for different applications such as person recognition, image segmentation, image captioning, scene description, and action recognition. In this dissertation, we study the deep learning models and their application in improving the performance and reliability of person recognition. This dissertation focuses on five aspects of person recognition: (1) multimodal person recognition, (2) quality-aware multi-sample person recognition, (3) text-independent speaker verification, (4) adversarial iris examples, and (5) morphed face images. First, we discuss the application of multimodal networks consisting of face, iris, fingerprint, and speech modalities in person recognition. We propose multi-stream convolutional neural network architectures …

Deep Learning Architectures For Heterogeneous Face Recognition, Seyed Mehdi Iranmanesh

Graduate Theses, Dissertations, and Problem Reports

Face recognition has been one of the most challenging areas of research in biometrics and computer vision. Many face recognition algorithms are designed to address illumination and pose problems for visible face images. In recent years, there has been significant amount of research in Heterogeneous Face Recognition (HFR). The large modality gap between faces captured in different spectrum as well as lack of training data makes heterogeneous face recognition (HFR) quite a challenging problem. In this work, we present different deep learning frameworks to address the problem of matching non-visible face photos against a gallery of visible faces.

Algorithms for …

Sensor Data Based Adaptive Models For Assembly Worker Training In Cyber Manufacturing, Md. Al-Amin

Doctoral Dissertations

“Production innovations are occurring faster than ever leading conventional production systems towards cyber manufacturing. Manufacturing workers thus need to frequently learn new methods and skills. In fast-changing, largely uncertain production systems, manufacturers with the ability to comprehend workers’ behavior and assess their operational performance in near real-time will achieve better performance than peers. Recognizing worker actions in near real-time while performing the assembly can serve this purpose. However, reliably recognizing the assembly actions performed by the workers is challenging, because the actions for assembly are complex and workers are not only heterogeneous but sensitive to the variation of the work …

A Novel Approach Toward High Accuracy Indoor Localization}{Masters Of Science, Yunshu Wang

Electronic Theses and Dissertations

This thesis presents a novel approach towards high accuracy indoor localization with smartphones. Better than all the existing indoor localization methods, our approach has the advantage of being infrastructural-free, robust, and it does not require any pre-installation in a new environment. To make this goal come true, we built a testbed that only uses the Inertial Measurement Units (IMU) of the smartphones to access the smartphones' raw acceleration and orientation data, then use these data to calculate the user's location by coordinate transformation and interactions. We conducted extensive experiments and evaluations for testbed validation as well as to carefully examine …

Leveraging Domain Knowledge In Deep Learning Systems, Colin Van Oort

Graduate College Dissertations and Theses

Machine learning, and the sub-field of deep learning in particular, has experienced an explosion in research interest and practical applications over the past few decades. Deep learning approaches seem to have become the preferred approach in many domains, outpacing the use of more traditional machine learning methods. This transitionhas also coincided with a shift away from feature engineering based on domain knowledge. Instead, the common deep learning philosophy is to learn relevant features through the combination of expressive models and large datasets. Some have interpreted this paradigm shift as the death of domain knowledge. I argue that domain knowledge is …

A New Distributed Anomaly Detection Approach For Log Ids Management Based Ondeep Learning, Murat Koca, Muhammed Ali̇ Aydin, Ahmet Sertbaş, Abdül Hali̇m Zai̇m

Turkish Journal of Electrical Engineering and Computer Sciences

Today, with the rapid increase of data, the security of big data has become more important than ever for managers. However, traditional infrastructure systems cannot cope with increasingly big data that is created like an avalanche. In addition, as the existing database systems increase licensing costs per transaction, organizations using information technologies are shifting to free and open source solutions. For this reason, we propose an anomaly attack detection model on Apache Hadoop distributed file system (HDFS), which stands out in open source big data analytics, and Apache Spark, which stands out with its speed performance in analysis to reduce …

Computational Intelligent Impact Force Modeling And Monitoring In Hislo Conditions For Maximizing Surface Mining Efficiency, Safety, And Health, Danish Ali

Doctoral Dissertations

"Shovel-truck systems are the most widely employed excavation and material handling systems for surface mining operations. During this process, a high-impact shovel loading operation (HISLO) produces large forces that cause extreme whole body vibrations (WBV) that can severely affect the safety and health of haul truck operators. Previously developed solutions have failed to produce satisfactory results as the vibrations at the truck operator seat still exceed the “Extremely Uncomfortable Limits”. This study was a novel effort in developing deep learning-based solution to the HISLO problem.

This research study developed a rigorous mathematical model and a 3D virtual simulation model to …