Engineering Commons^™

Data-Driven Vibration-Based Condition Monitoring: Fundamentals, Applications, And Challenges, Sulaiman A. S. Aburakhia

Electronic Thesis and Dissertation Repository

Vibration-Based Condition Monitoring (VBCM) is commonly utilized in Prognostics and Health Management (PHM) due to its non-destructive nature and inherent advantages over alternative forms of condition monitoring. Furthermore, the rapid evolution of sensor fabrication and the rise of the Internet of Things (IoT) have facilitated large-scale VBCM systems across diverse domains, including industry, transportation, healthcare, agriculture, and wildlife monitoring. The recent advancements in computing technologies have significantly expanded the potential for VBCM by leveraging the synergy between signal processing and Machine Learning (ML). Accordingly, data-driven VBCM has emerged as a paradigm shift, improving the performance and reliability of VBCM systems. …

Feature Extraction And Design In Deep Learning Models, Daniel Perez

Computational Modeling & Simulation Engineering Theses & Dissertations

The selection and computation of meaningful features is critical for developing good deep learning methods. This dissertation demonstrates how focusing on this process can significantly improve the results of learning-based approaches. Specifically, this dissertation presents a series of different studies in which feature extraction and design was a significant factor for obtaining effective results. The first two studies are a content-based image retrieval system (CBIR) and a seagrass quantification study in which deep learning models were used to extract meaningful high-level features that significantly increased the performance of the approaches. Secondly, a method for change detection is proposed where the …

A Scale Space Local Binary Pattern (Sslbp) – Based Feature Extraction Framework To Detect Bones From Knee Mri Scans, Jinyeong Mun

Electronic Theses and Dissertations

The medical industry is currently working on a fully autonomous surgical system, which is considered a novel modality to go beyond technical limitations of conventional surgery. In order to apply an autonomous surgical system to knees, one of the primarily responsible areas for supporting the total weight of human body, accurate segmentation of bones from knee Magnetic Resonance Imaging (MRI) scans plays a crucial role. In this paper, we propose employing the Scale Space Local Binary Pattern (SSLBP) feature extraction, a variant of local binary pattern extractions, for detecting bones from knee images. The proposed methods consist of two phases. …

Feasibility Of Melville Marginalia Authorship Differentiation, Aaron Burdin

Boise State University Theses and Dissertations

We examine the feasibility of using image processing techniques to determine differentiation in authorship of historical pencil marks. Pencil marks with unattributed and attributed authorship are segmented from digital images of historical books. Analysis is performed on five features that are extracted from the "vertical" pencil marks, with those features used as a basis for authorship of marks. These marks consist of single stroke marks that are interspersed in the same document. We describe the challenges of the digital format that we were given and the steps taken in using autonomous segmentation to save pixel locations of marks. Five mark …

Using Computer Vision To Build A Predictive Model Of Fruit Shelf-Life, Nandan G. Thor

Master's Theses

Computer vision is becoming a ubiquitous technology in many industries on account of its speed, accuracy, and long-term cost efficacy. The ability of a computer vision system to quickly and efficiently make quality decisions has made computer vision a popular technology on inspection lines. However, few companies in the agriculture industry use computer vision because of the non-uniformity of sellable produce. The small number of agriculture companies that do utilize computer vision use it to extract features for size sorting or for a binary grading system: if the piece of fruit has a certain color, certain shape, and certain size, …

Geometric Accuracy Evaluation Of Mobile Terrestrial Lidar Surveys With Supporting Algorithms, Chisaphat Supunyachotsakul

Open Access Dissertations

Mobile Mapping System (MMS) technology is widely used for many applications, hence quantifying its accuracy is a very important and essential task and is a primary focus of this research. In general, to perfrom geometric accuracy evaluation of MMS data, validation points/features are needed. A method is needed to capture a point feature off the roadway in a position where a target on the ground surface would not be visible to the scanner. In this study, eight sphere targets with 14" diameter were placed on the shoulder of the roadway over validation points on the ground. The sphere targets were …

Preprocessing Techniques To Support Event Detection Data Fusion On Social Media Data, Brandon T. Davis

Theses and Dissertations

This thesis focuses on collection and preprocessing of streaming social media feeds for metadata as well as the visual and textual information. Today, news media has been the main source of immediate news events, large and small. However, the information conveyed on these news sources is delayed due to the lack of proximity and general knowledge of the event. Such news have started relying on social media sources for initial knowledge of these events. Previous works focused on captured textual data from social media as a data source to detect events. This preprocessing framework postures to facilitate the data fusion …

Sparse Coding Based Feature Representation Method For Remote Sensing Images, Ender Oguslu

Electrical & Computer Engineering Theses & Dissertations

In this dissertation, we study sparse coding based feature representation method for the classification of multispectral and hyperspectral images (HSI). The existing feature representation systems based on the sparse signal model are computationally expensive, requiring to solve a convex optimization problem to learn a dictionary. A sparse coding feature representation framework for the classification of HSI is presented that alleviates the complexity of sparse coding through sub-band construction, dictionary learning, and encoding steps. In the framework, we construct the dictionary based upon the extracted sub-bands from the spectral representation of a pixel. In the encoding step, we utilize a soft …

Medical Image Registration Using Artificial Neural Network, Hyunjong Choi

Master's Theses

Image registration is the transformation of different sets of images into one coordinate system in order to align and overlay multiple images. Image registration is used in many fields such as medical imaging, remote sensing, and computer vision. It is very important in medical research, where multiple images are acquired from different sensors at various points in time. This allows doctors to monitor the effects of treatments on patients in a certain region of interest over time. In this thesis, artificial neural networks with curvelet keypoints are used to estimate the parameters of registration. Simulations show that the curvelet keypoints …

Identifying Image Manipulation Software From Image Features, Devlin T. Boyter

Theses and Dissertations

As technology steadily increases in the field of image manipulation, determining which software was used to manipulate an image becomes increasingly complex for law enforcement and intelligence agencies. To combat this difficult problem, new techniques that examine the artifacts left behind by a specific manipulation are converted to features for classification. This research implemented four preexisting image manipulation detection techniques into a framework of modules: Two-Dimensional Second Derivative, One-Dimensional Zero Crossings, Quantization Matrices Identification, and File Metadata analysis. The intent is the creation of a framework to develop a capability to determine which specific image manipulation software program manipulated an …

Feature Extraction And Recognition For Human Action Recognition, Jiajia Luo

Doctoral Dissertations

How to automatically label videos containing human motions is the task of human action recognition. Traditional human action recognition algorithms use the RGB videos as input, and it is a challenging task because of the large intra-class variations of actions, cluttered background, possible camera movement, and illumination variations. Recently, the introduction of cost-effective depth cameras provides a new possibility to address difficult issues. However, it also brings new challenges such as noisy depth maps and time alignment. In this dissertation, effective and computationally efficient feature extraction and recognition algorithms are proposed for human action recognition.

At the feature extraction step, …

Analyses Of Online Monitoring Signals For A Gmaw Process Before And After Improvement, Aniruddha Joshi

LSU Master's Theses

The ability to detect the onset of welding instability is a very powerful tool in welding process monitoring and control. Toward this goal, this study investigates a gas metal arc welding (GMAW) process by analyzing online monitoring signals. Two separate data sets are obtained from the process, which correspond to (a) a stable process after improvement and (b) a relatively unstable process which tends to exhibit spatter and poor weld bead geometry. Voltage, current, wire-feeding speed and line speed signals for both data sets are analysed and features are extracted from the raw signals using different signal processing techniques. Specifically, …

Cervical Cancer Histology Image Feature Extraction And Classification, Peng Guo

Masters Theses

"Cervical cancer, the second most common cancer affecting women worldwide and the most common in developing countries can be cured if detected early and treated. Expert pathologists routinely visually examine histology slides for cervix tissue abnormality assessment. In previous research, an automated, localized, fusion-based approach was investigated for classifying squamous epithelium into Normal, CIN1, CIN2, and CIN3 grades of cervical intraepithelial neoplasia (CIN) based on image analysis of 62 digitized histology images obtained through the National Library of Medicine. In this research, CIN grade assessments from two pathologists are analyzed and are used to facilitate atypical cell concentration feature development …

Vocal Fold Analysis From High Speed Videoendoscopic Data, Jing Chen

LSU Doctoral Dissertations

High speed videoendoscopy (HSV) of the larynx far surpasses the limits of videostroboscopy in evaluating the vocal fold vibratory behavior by providing much higher frame rate. HSV enables the visualization of vocal fold vibratory pattern within an actual glottic cycle. This very detailed infor-mation on vocal fold vibratory characteristics could provide valuable information for the assessment of vocal fold vibratory function in disordered voices and the treatments effects of the behavioral, medical and surgical treatment procedures. In this work, we aim at addressing the problem of classi-fying voice disorders with varying etiology by following four steps described shortly. Our method-ology …

Experiments With Gmti Radar Using Micro-Doppler, Benjamin Walter Dilsaver

Theses and Dissertations

As objects move, their changing shape produces a signature that can be measured by a radar system. That signature is called the micro-Doppler signature. The micro-Doppler signature of an object is a distinguishing characteristic for certain classes of objects. In this thesis features are extracted from the micro-Doppler signature and are used to classify objects. The scope of the objects is limited to humans walking and traveling vehicles. The micro-Doppler features are able to distinguish the two classes of objects. With a sufficient amount of training data, the micro-Doppler features may be used with learning algorithms to predict unknown objects …

Feature Extraction Through K-Means Segmentation For Melanoma Detection, Snigdha Priya Bommadevara

Masters Theses

"Malignant melanoma is responsible for 75% of the deaths caused due to skin cancer annually. However, melanoma detection can be possible through feature extraction and pattern classification, which can lower the risk, if the melanoma is detected at an early stage. Clustering is one of the most useful tools used to differentiate features that can contribute to melanoma. This research work uses the k-means clustering algorithm for implementation of color segmentation. However, k-means clustering requires a predefined value of k, i.e., the number of clusters must be specified at the beginning of the run. This research uses a predefined value …

Structural Health Monitoring Of Historic Masonry Monuments, Saurabh Prabhu

All Theses

Structural Health Monitoring (SHM) is a well-accepted diagnostic technique being used to evaluate modern structures. This method involves monitoring the vibration response of a structure to detect changes in its structural state. The primary intention of this thesis is to address two practical and technical difficulties encountered in deploying SHM on historic masonry monuments: (i) the selection of suitable low dimensional vibration response features that are highly sensitive to the presence and extent of damage, while having low sensitivity to extraneous noise and (ii) the selection of optimal sensor locations for efficient system identification applied to Gothic Cathedrals. Both of …

Bemdec: An Adaptive And Robust Methodology For Digital Image Feature Extraction, Isaac Kueth Gang

Dissertations

The intriguing study of feature extraction, and edge detection in particular, has, as a result of the increased use of imagery, drawn even more attention not just from the field of computer science but also from a variety of scientific fields. However, various challenges surrounding the formulation of feature extraction operator, particularly of edges, which is capable of satisfying the necessary properties of low probability of error (i.e., failure of marking true edges), accuracy, and consistent response to a single edge, continue to persist. Moreover, it should be pointed out that most of the work in the area of feature …

A Computational Fluid Dynamics Feature Extraction Method Using Subjective Logic, Clifton H. Mortensen

Theses and Dissertations

Computational fluid dynamics simulations are advancing to correctly simulate highly complex fluid flow problems that can require weeks of computation on expensive high performance clusters. These simulations can generate terabytes of data and pose a severe challenge to a researcher analyzing the data. Presented in this document is a general method to extract computational fluid dynamics flow features concurrent with a simulation and as a post-processing step to drastically reduce researcher post-processing time. This general method uses software agents governed by subjective logic to make decisions about extracted features in converging and converged data sets. The software agents are designed …

Computer-Aided Detection Of Sleep Apnea And Sleep Stage Classification Using Hrv And Eeg Signals, Edson F. Estrada

Open Access Theses & Dissertations

Sleep is a circadian rhythm essential for human life. Many events occur in the body during this state. In the past, significant efforts have been made to provide clinicians with reliable and less intrusive tools to automatically classify the sleep stages and detect apnea events. A few systems are available in the market to accomplish this task. However, sleep specialists may not have full confidence and trust in such systems due to issues related to their accuracy, sensitivity and specificity. The main objective of this work is to explore possible relationships among sleep stages and apneic events and improve on …

Improved Feature Extraction, Feature Selection, And Identification Techniques That Create A Fast Unsupervised Hyperspectral Target Detection Algorithm, Robert J. Johnson

Theses and Dissertations

This research extends the emerging field of hyperspectral image (HSI) target detectors that assume a global linear mixture model (LMM) of HSI and employ independent component analysis (ICA) to unmix HSI images. Via new techniques to fully automate feature extraction, feature selection, and target pixel identification, an autonomous global anomaly detector, AutoGAD, has been developed for potential employment in an operational environment for real-time processing of HSI targets. For dimensionality reduction (initial feature extraction prior to ICA), a geometric solution that effectively approximates the number of distinct spectral signals is presented. The solution is based on the theory of the …

Hybrid And Hierarchical Image Registration Techniques, Dongjiang Xu

Electronic Theses and Dissertations

A large number of image registration techniques have been developed for various types of sensors and applications, with the aim to improve the accuracy, computational complexity, generality, and robustness. They can be broadly classified into two categories: intensity-based and feature-based methods. The primary drawback of the intensity-based approaches is that it may fail unless the two images are misaligned by a moderate difference in scale, rotation, and translation. In addition, intensity-based methods lack the robustness in the presence of non-spatial distortions due to different imaging conditions between images. In this dissertation, the image registration is formulated as a two-stage hybrid …

Signal Modeling With Non-Uniform Time Sampling Of Features For Automatic Speech Recognition, Montri Karnjanadecha

Electrical & Computer Engineering Theses & Dissertations

This dissertation presents an investigation of non-uniform time sampling methods for spectral/temporal feature extraction in speech. Frame-based features were computed based on an encoding of the global spectral shape using a Discrete Cosine Transform. In most current “standard” methods, trajectory (dynamic) features are determined from frame-based parameters using a fixed time sampling, i.e., fixed block length and fixed block spacing. In this research, new methods are proposed and investigated in which block length and/or block spacing are variable. The idea was initially tested with HMM-based isolated word recognition, and a significant performance improvement resulted when a variable block length and …

Real Time Detection Of Anomalous Satellite Behavior Has Ground Based Telescope Images, Geoffrey S. Maron

Theses and Dissertations

Air Force analysts are faced with the task of monitoring satellites with ground based telescopes. Images are collected and analyzed in a time consuming and subjective effort to detect any behavior that is anomalous. This research maximizes use of a priori information to create an automated, real time satellite behavior classification tool. Using modeling software and knowledge of a satellite's orbit, reference imagery is created for each measured image in a satellite pass. Features are extracted from the measured and reference image pairs that provide good overall Gaussian classification accuracy (85%), reduce the dimensionality of the problem (from 32,768 down …