Digital Commons Network^™

Threshold Free Detection Of Elliptical Landmarks Using Machine Learning, Lifan Zhang

Theses and Dissertations

Elliptical shape detection is widely used in practical applications. Nearly all classical ellipse detection algorithms require some form of threshold, which can be a major cause of detection failure, especially in the challenging case of Moire Phase Tracking (MPT) target images. To meet the challenge, a threshold free detection algorithm for elliptical landmarks is proposed in this thesis. The proposed Aligned Gradient and Unaligned Gradient (AGUG) algorithm is a Support Vector Machine (SVM)-based classification algorithm, original features are extracted from the gradient information corresponding to the sampled pixels. with proper selection of features, the proposed algorithm has a high accuracy …

An Analog Cmos Particle Filter, Trevor Watson

Masters Theses

Particle filters are used in a variety of image processing and machine learning applications. Their main use in these applications is to gather information about a system of objects, by using partial or noisy observations collected from sensors. These observations are used to associate points of interest in the observations with objects and maintain this association through a series of observations.

In this paper I will investigate the performance of a particle filter implemented in 130nm analog CMOS hardware. The design goal of the particle filter is low-microwatt power consumption. Using analog hardware, rather than digital ASICs or CPUs I …

Graph-Based Latent Embedding, Annotation And Representation Learning In Neural Networks For Semi-Supervised And Unsupervised Settings, Ismail Ozsel Kilinc

USF Tampa Graduate Theses and Dissertations

Machine learning has been immensely successful in supervised learning with outstanding examples in major industrial applications such as voice and image recognition. Following these developments, the most recent research has now begun to focus primarily on algorithms which can exploit very large sets of unlabeled examples to reduce the amount of manually labeled data required for existing models to perform well. In this dissertation, we propose graph-based latent embedding/annotation/representation learning techniques in neural networks tailored for semi-supervised and unsupervised learning problems. Specifically, we propose a novel regularization technique called Graph-based Activity Regularization (GAR) and a novel output layer modification called …

Adaft: A Resource-Efficient Framework For Adaptive Fault-Tolerance In Cyber-Physical Systems, Ye Xu

Doctoral Dissertations

Cyber-physical systems frequently have to use massive redundancy to meet application requirements for high reliability. While such redundancy is required, it can be activated adaptively, based on the current state of the controlled plant. Most of the time the physical plant is in a state that allows for a lower level of fault-tolerance. Avoiding the continuous deployment of massive fault-tolerance will greatly reduce the workload of CPSs. In this dissertation, we demonstrate a software simulation framework (AdaFT) that can automatically generate the sub-spaces within which our adaptive fault-tolerance can be applied. We also show the theoretical benefits of AdaFT, and …

Machine Learning Based Method Of Moments (Ml-Mom), He Ming Yao, Li (Lijun) Jun Jiang, Yu Wei Qin

Electrical and Computer Engineering Faculty Research & Creative Works

This paper proposes a novel method by rethinking the method of moments (MoM) solving process into a machine learning training process. Based on the artificial neural network (ANN), the conventional MoM matrix is treated as the training data set, based on which machine learning training process becomes conventional linear algebra MoM solving process. The trained result is the solution of MoM. The multiple linear regression (MLR) is utilized to train the model. Amazon Web Service (AWS) is used as the computations platform to utilize the existing hardware and software resources for machine learning. To verify the feasibility of the proposed …

Information Theoretic Study Of Gaussian Graphical Models And Their Applications, Ali Moharrer

LSU Doctoral Dissertations

In many problems we are dealing with characterizing a behavior of a complex stochastic system or its response to a set of particular inputs. Such problems span over several topics such as machine learning, complex networks, e.g., social or communication networks; biology, etc. Probabilistic graphical models (PGMs) are powerful tools that offer a compact modeling of complex systems. They are designed to capture the random behavior, i.e., the joint distribution of the system to the best possible accuracy. Our goal is to study certain algebraic and topological properties of a special class of graphical models, known as Gaussian graphs. First, …

Bayesian Methods And Machine Learning For Processing Text And Image Data, Yingying Gu

Theses and Dissertations

Classification/clustering is an important class of unstructured data processing problems. The classification (supervised, semi-supervised and unsupervised) aims to discover the clusters and group the similar data into categories for information organization and knowledge discovery. My work focuses on using the Bayesian methods and machine learning techniques to classify the free-text and image data, and address how to overcome the limitations of the traditional methods. The Bayesian approach provides a way to allow using more variations(numerical or categorical), and estimate the probabilities instead of explicit rules, which will benefit in the ambiguous cases. The MAP(maximum a posterior) estimation is used to …

Prediction Of Graduation Delay Based On Student Characterisitics And Performance, Tushar Ojha

Electrical and Computer Engineering ETDs

A college student's success depends on many factors including pre-university characteristics and university student support services. Student graduation rates are often used as an objective metric to measure institutional effectiveness. This work studies the impact of such factors on graduation rates, with a particular focus on delay in graduation. In this work, we used feature selection methods to identify a subset of the pre-institutional features with the highest discriminative power. In particular, Forward Selection with Linear Regression, Backward Elimination with Linear Regression, and Lasso Regression were applied. The feature sets were selected in a multivariate fashion. High school GPA, ACT …

Predictive Analytics In Cardiac Healthcare And 5g Cellular Networks, Dilranjan S. Wickramasuriya

USF Tampa Graduate Theses and Dissertations

This thesis proposes the use of Machine Learning (ML) to two very distinct, yet compelling, applications – predicting cardiac arrhythmia episodes and predicting base station association in 5G networks comprising of virtual cells. In the first scenario, Support Vector Machines (SVMs) are used to classify features extracted from electrocardiogram (EKG) signals. The second problem requires a different formulation departing from traditional ML classification where the objective is to partition feature space into constituent class regions. Instead, the intention here is to identify temporal patterns in unequal-length sequences. Using Recurrent Neural Networks (RNNs), it is demonstrated that accurate predictions can be …

Machine Learning Based Mom (Ml-Mom) For Parasitic Capacitance Extractions, He Ming Yao, Yu Wei Qin, Li (Lijun) Jun Jiang

Electrical and Computer Engineering Faculty Research & Creative Works

This paper is a rethinking of the conventional method of moments (MoM) using the modern machine learning (ML) technology. By repositioning the MoM matrix and unknowns in an artificial neural network (ANN), the conventional linear algebra MoM solving is changed into a machine learning training process. The trained result is the solution. As an application, the parasitic capacitance extraction broadly needed by VLSI modeling is solved through the proposed new machine learning based method of moments (ML-MoM). The multiple linear regression (MLR) is employed to train the model. The computations are done on Amazon Web Service (AWS). Benchmarks demonstrated the …

Explorations Into Machine Learning Techniques For Precipitation Nowcasting, Aditya Nagarajan

Masters Theses

Recent advances in cloud-based big-data technologies now makes data driven solutions feasible for increasing numbers of scientific computing applications. One such data driven solution approach is machine learning where patterns in large data sets are brought to the surface by finding complex mathematical relationships within the data. Nowcasting or short-term prediction of rainfall in a given region is an important problem in meteorology. In this thesis we explore the nowcasting problem through a data driven approach by formulating it as a machine learning problem.

State-of-the-art nowcasting systems today are based on numerical models which describe the physical processes leading to …

Quantitative Estimation Of Causality And Predictive Modeling For Precipitation Observation Sites And River Gage Sensors, Tri Vu Nguyen

LSU Master's Theses

This project seeks to investigate two questions: correlations from precipitation measurement sensors to river gage sensors, and predictive modeling of peak river gage heights during precipitation events. First, if correlations can be quantified, then a predictive model can be explored to predict peak water levels at river gage sensors, in response to precipitation inputs. Answering both research questions can provide early flood detection benefits and provide quantitative time assessments for flood risks. An extensive data-driven study was conducted across a geographical area of the U.S, spanning the time period 2008-2016 to identify river gage sensors that are closely correlated to …