Doctor of Philosophy, The Ohio State University, 2021, Electrical and Computer Engineering

As communication and control systems become more complex, connected and process data at higher velocities, detecting changes in patterns becomes increasingly difficult yet still crucial to guarantee a level of QoS, security, reliability etc. For large systems, usually there are many modes of failure and they are also prone to attacks from different surfaces. Still, there are numerous zero-day vulnerabilities that are unidentified until they cause a fault or are exploited. Change detection with unknown distributions provides a way of detecting the occurrence of faults or the gain of access by malicious parties by comparing the time series system features to their norm. In a wide variety of the applications, on the other hand, it is feasible to assume a certain level of knowledge of the system before the effect takes place and utilizing the knowledge of initial conditions increases the detection performance. With an ever increasing data rate and connectivity, any change in the observed process has to be detected on the fly before it is outdated, without the necessity to store and with a small blast radius for malicious activities. A delay in real time change detection may result in QoS disruption, cyber-physical threats and inability to contain the spread of a disease. So, minimal computational complexity is a key ingredient for modern change detection algorithms. In this dissertation, we assume non-Bayesian change detection problems under a finite alphabet with varying change point and cost models and with unknown post-change distributions. We focus on robust detection algorithms that utilize the knowledge of pre-change system dynamics and are of low complexity. Given that the effect of the change on the system is unknown, the distribution of observations may divert in many ways without much structure, whereas, before the change point, a false alarm is structured by Sanov's theorem, following a particular sample path. The proposed methods characterize (open full item for complete abstract)

Committee: C. Emre Koksal (Advisor); Atilla Eryilmaz (Committee Member); Kiryung Lee (Committee Member); Abhishek Gupta (Committee Member) Subjects: Computer Engineering; Electrical Engineering; Statistics

PHD, Kent State University, 2020, College of Arts and Sciences / Department of Mathematical Sciences

High-dimensional offline and online time series data are characterized by a large number of measurements and complex dependence, and often involve change points. Change point detection in offline time series data improves the parameter testing and estimation by pooling homogeneous observations between two successive change points. Change point detection in online time series data provides timely snapshots of the monitored system and allows for real-time anomaly detection. Despite its importance, methods available for change point detection in high-dimensional offline and online time series data are scarce. In the first part of the thesis, we present some new statistics for change-point testing and estimation in high dimensional offline time series data. We establish their theoretical properties including asymptotic distributions and consistency under mild conditions. The developed new methods are non-parametric without imposing restrictive structural assumptions. They incorporate spatial and temporal dependence in data. Most importantly, they can detect the change point near the boundary of time series data. In the second part of the thesis, we extend these new statistics to high-dimensional online time series data and provide a new stopping rule to detect a change point as early as possible after an anomaly occurs. We study theoretical properties of the new stopping rule, and derive an explicit expression for the average run length (ARL) so that the level of threshold in the stopping rule can be easily obtained with no need to run time-consuming Monte Carlo simulations. We also establish an upper bound for the expected detection delay (EDD), which demonstrates the impact of data dependence and magnitude of structure change in data. Simulation and case studies are provided to demonstrate the empirical performance of the proposed offline and online change-point detection methods.

Committee: Jun Li (Advisor); Mohammad Khan (Committee Member); Jing Li (Committee Member); Cheng-Chang Lu (Committee Member); Ruoming Jin (Committee Member) Subjects: Mathematics; Statistics

Doctor of Philosophy, The Ohio State University, 2013, Electrical and Computer Engineering

Until recently, the functionality of radar systems has been built into the radar's analog hardware, resulting in radars which are inflexible and that can only be used for a specific application. Modern systems, however, driven by the ever increasing speed of processors and data converters - analog-to-digital (ADC) and digital-to-analog (DAC) - are transitioning toward software defined radar (SDR) systems. The advent of SDRs inevitably leads to the question of how their added flexibilities can best be leveraged. The work within this dissertation is motivated by joint radar and communication functionality. The main objective is to study and demonstrate the ability of radar systems to employ non-traditional, specifically, communication waveforms for remote sensing. A software defined radar (SDR) is developed. The SDR features a "closed loop" testbed interface accessible via Matlab m-code. Here, "closed-loop" means that data can be pulled from the SDR, processed, then used to select/adapt the waveform and settings of the SDR without human intervention, i.e. on the fly. The testbed interface is used to implement a joint radar and communication system which is capable of collecting and processing radar data, e.g. range-Doppler maps, while simultaneously communicating previously collected radar data. Simultaneous functionality is accomplished by interrogating with a wide band digital communication waveform which is modulated with the previously collected radar data. The joint system is used to empirically demonstrate the theoretical work on detection and change detection within this dissertation. Optimal detectors are developed for interrogation with communication waveforms. The optimal detector for a single target with known impulse response in white noise is known to be a thresholding of the output of a matched filter. Radar systems, however, often operate in multi-target environments; notably air-to-ground synthetic aperture radars. For such applic (open full item for complete abstract)

Committee: Emre Ertin (Advisor); Randolph Moses (Advisor); Chris Baker (Committee Member) Subjects: Electrical Engineering

PhD, University of Cincinnati, 2017, Allied Health Sciences: Communication Sciences and Disorders

Objective: The acoustic change complex (ACC) is a type of cortical auditory evoked potentials (CAEP) elicited by an acoustic change in an ongoing stimulus. The purpose of this study were: 1) to characterize waveform features and brain activation patterns of the ACC evoked by frequency changes, and 2) to determine if the ACC can serve as an objective tool to predict frequency change detection threshold (FCDT) measured behaviorally in cochlear implant (CI) users. Methods: Twelve post-lingually deafened adult CI users and 12 normal-hearing (NH) young listeners participated in this study. A psychoacoustic test and electroencephalography (EEG) recordings were administered in each participant. For the psychoacoustic test, the stimuli were a series of 160 Hz base frequency tones containing different magnitudes of upward frequency changes. The FCDTs were measured. For EEG recordings, the stimuli were similar to those in the psychoacoustic test. The base frequencies were 160 Hz or 1200 Hz. The magnitudes of frequency changes were 0% (no change), 5%, and 50% change at each base frequency. The onset CAEP (N1-P2 complex) evoked by stimulus onset and the ACC (N1'-P2' complex) evoked by frequency changes were analyzed. A source localization analysis with standardized low-resolution brain electromagnetic tomography (sLORETA) was further used to localize brain regions that were associated with the onset CAEP N1 and the ACC N1'. The correlation between the ACC measures (wave peak amplitude, latency, and current source density) and the FCDT was examined. Results: The ACC and onset CAEP peaks displayed longer latencies in CI users than in NH listeners (p<0.05), with this trend more prominent for the ACC peaks. The N1'P2 amplitude of the ACC was significantly smaller in CI users than in NH listeners (p<0.05). With the 160 Hz base frequency, the N1' latency evoked by 50% frequency change was significantly correlated with the FCDT in CI group (r=0.48, p<0.05). The current source (open full item for complete abstract)

Committee: Fawen Zhang Ph.D. (Committee Chair); Pamara Chang (Committee Member); Brian Earl (Committee Member); Noah Silbert Ph.D. (Committee Member) Subjects: Health Sciences

Master of Science (MS), Wright State University, 2003, Geological Sciences

Hurley, Angela Lorraine. M.S., Department of Geological Sciences, Wright State University, 2003. Identification of Gypsy Moth Defoliation in Ohio Using Landsat Data. The gypsy moth is one of the most devastating forest pests in North America. In late spring, gypsy moth larvae hatch from eggs laid the previous summer. During the next forty days, tens of thousands of these caterpillars eat up to one square foot of foliage each. The gypsy moth has established populations in several states, and dangerously fast-growing populations in several others. The state of Ohio is a critical area in the suppression of the gypsy moth because the front of gypsy moth advance passes through the state. Besides diminishing the aesthetic value of Ohio's forests, gypsy moths also cause substantial economic damage to the Ohio timber industry, which is estimated to be a $7 billion per year industry. The Ohio Department of Agriculture currently uses aerial sketchmapping each year to assess the damage done by the gypsy moth. This procedure is difficult, time-consuming, and somewhat imprecise. The results obtained from Landsat 5 and Landsat 7 data can be compared to locations determined by aerial sketchmapping to locate gypsy moth infestations in Ohio. Since vegetation reflects infrared light and absorbs visible light, the health of vegetation can be assessed using a haze-adjusted ratio of Landsat spectral band 4 (near-infrared) to Landsat spectral band 3 (visible red). To determine the change that has occurred between two dates, the ratio values from two dates are subtracted. To identify change that has been caused by the gypsy moth, an area should exhibit defoliation between early June and late June and subsequent refoliation between late June and late July. This type of change results in large positive ratio subtraction values between early June and late June and large negative ratio subtraction values between late June and late July. Pixels that exhibit these attributes are candidates for (open full item for complete abstract)

Committee: Doyle Watts (Advisor) Subjects: Geology

Master of Science (MS), Ohio University, 2004, Environmental Studies (Arts and Sciences)

In Belize, the use of remotely sensed information for monitoring landscape dynamics is a relatively new area. This study takes advantage of contemporary technologies, such as remote sensing, for monitoring land use and land cover changes in Belize. The study area covers approximately 6,190 square miles. Two Landsat images of 1993 and 2003 were used to identify, quantify, assess and map changes in land use and land cover. The Landsat images were classified using an unsupervised K-means algorithm. Comparison of ground truth points and the 2003 classification result shows a classification accuracy of 92%. The digital change detection methodology involved a pixel-by-pixel comparison of the classified images using ENVI software. The results show that urban expansion (12%/year) is occurring at a faster rate than population growth (3.5%/year). In addition, agricultural land expansion is occurring at a rate of 32 square miles per annum. Urban development, agricultural land expansion and extensive pine forest cover loss are contributing to an estimated deforestation rate of 35 square miles per annum. In general, this study provides urgent and needed information that will guide the Government of Belize to achieve the desired goals of sustainable development.

Committee: James Lein (Advisor) Subjects: Environmental Sciences

Master of Arts, Miami University, 2007, Geography

The main goal of the study was mapping land cover land use change patterns in Mbeere District between 1987 and 2000.Two Landsat images acquired in 1987 and 2000; and MODIS data were used to map and quantify the patterns of change. The results revealed a complex land cover change pattern between the two dates; with both positive and negative changes. Grasslands increased by 29 %, settlement/agriculture by 31 %, while woodland reduced by 41%. The study also confirmed that digital change detection is still a viable change detection method in arid and semi-arid lands despite limitations associated with factors like high spectral similarities and phenology.

Committee: Mary Henry (Advisor) Subjects:

Master of Science, The Ohio State University, 2023, Civil Engineering

Recent advances in satellite imaging have seen the thriving of PlanetScope data. A leading feature of this data is the high temporal frequency at which imagery is captured, although at a low to medium spatial resolution. Additionally, these images are often accompanied by cloud masks and other auxiliary information, further enhancing their usability. The simplistic access provided by PlanetScope's online portal emphasizes its popularity. Such advancements naturally lead to inquiries about the efficiency of PlanetScope data in change detection analysis. The principal aim of this investigation is to evaluate the suitability and efficacy of PlanetScope imagery when applied to established change detection models, with the ultimate goal of determining the utility of PlanetScope data for the purpose of change detection. For evaluation purposes, the chosen models were divided into two categories: those based on traditional change detection methodologies, and those leveraging advanced deep learning algorithms. This research utilized three previously developed models, applying them to the PlanetScope dataset to ascertain its adaptability and effectiveness. These models were subsequently juxtaposed for comparative analysis. The initial model was influenced by [Kondmann et al., 2021], which developed a half-sibling regression technique founded on the principles proposed by [Scholkopf et al., 2016]. This model was subsequently termed the Sibling Regression for Optical Change Detection (SiROC). In contrast, the latter two models are grounded in deep learning models. Both models incorporate the Siamese network architecture; one utilizes it in conjunction with a fully convolutional network as outlined by [Daudt et al., 2018a], while the other pairs it with the Transformer architecture, drawing inspiration from [Bandara and Patel, 2022]. Upon analyzing the results derived from the simulation PlanetScope dataset, it was perceived that certain models experienced (open full item for complete abstract)

Committee: Rongjun Qin (Advisor); Wang Lei (Committee Member); Charles Toth (Committee Member) Subjects: Civil Engineering

Doctor of Philosophy, The Ohio State University, 2023, Environmental Science

Land cover changes refer to any modifications to the physical and biological components of an ecosystem and hence causing changes in its structure and functioning. Such changes encompass a range of magnitudes, from drastic changes such as urbanization and widespread deforestation, to subtle modifications such as forest disturbances, and even subtle shifts in the composition of crops and forest species. Land cover changes influence our climate system via two main pathways: the biogeochemical pathway (i.e., modifications to greenhouse gases), and the biophysical pathway (i.e., changes to land surface biophysics). However, analyzing the impact of land cover changes is challenging in part due to uncertainties of mapping the subtle changes, and also because a direct comparison between resultant biophysical and biogeochemical forcings is debatable as the effect of biophysical forcing is instant while the effect of biogeochemical forcing is long-lasting. The overarching goal of my research is to investigate the climate impact of subtle land cover changes and to build a framework that can compare biophysical impacts with corresponding biogeochemical impacts. To achieve this goal, three objectives are accomplished: (1) develop a multivariate model to better detect subtle land cover changes, (2) investigate the underlying biophysical mechanism regulating the post-change local climate, (3) incorporate the biophysical impact into the well-studied biogeochemical (carbon-centric) analyzing framework. Consistent findings following these objectives demonstrate that (1) the multivariate model we developed, which detects changes both in trends and seasonality, shows improved accuracy in detecting subtle land cover change (e.g., forest disturbance and following recoveries); (2) post-disturbance warming has been observed in all the climate zones except for the alpine tundra. Such warming is dominated by decreased-evapotranspiration-induced warming, even though offset by increased-a (open full item for complete abstract)

Committee: Kaiguang Zhao (Advisor); Yongyang Cai (Committee Member); Yanlan Liu (Committee Member); Gil Bohrer (Committee Member) Subjects: Climate Change; Environmental Economics; Environmental Science; Remote Sensing

Master of Science in Electrical Engineering (MSEE), Wright State University, 2018, Electrical Engineering

In this thesis, we develop and evaluate change detection algorithms for longwave infrared (LWIR) hyperspectral imagery. Because measured radiance in the LWIR domain depends on unknown surface temperature, care must be taken to prevent false alarms resulting from in-scene temperature differences that appear as material changes. We consider four strategies to mitigate this effect. In the first, pre-processing via traditional temperatureemissivity separation yields approximately temperature-invariant emissivity vectors for use in change detection. In the second, we utilize alpha residuals to obtain robustness to temperature errors. Next, we adopt a minimax approach that minimizes the maximal spectral deviation between measurements. Finally, we reduce our minmax approach to solve with fewer variables. Examples using synthetic and measured data quantify the computational complexity of the proposed methods and demonstrate orders of magnitude reduction in false alarm rates relative to existing methods.

Committee: Joshua Ash Ph.D. (Advisor); Fred Garber Ph.D. (Committee Member); Arnab Shaw Ph.D. (Committee Member) Subjects: Electrical Engineering

Doctor of Philosophy, University of Akron, 2018, Biomedical Engineering

Neurodegenerative diseases that are caused by deterioration of nerve cells in the brain and spinal cord affect more than 6 million Americans and cost nearly 0.8 trillion dollars annually in patient care. With a growing number of elderly population, the statistics are expected to worsen as there is currently no cure for these disorders. Modern medicines are at best palliative and only manage the symptoms. Therapeutic interventions to deliver functional neural cells to the ravaged tissue are essential to restore lost tissue functions. The use of stem cell-derived neural cells is a promising strategy for cell replacement therapies of neurodegenerative diseases. Embryonic stem cells (ESCs) are promising cell sources for therapeutic uses including cell replacement therapy of neural tissues. This is because ESCs have unlimited self-renewal and proliferation capabilities and the ability to differentiate into various neural cells. Nevertheless, despite significant investment and research, therapeutic uses of ESCs for neural cell replacement has been largely unsuccessful. Low and inconsistent yield of neural cells from ESCs and lack of a complete understanding of molecular mechanisms of neural differentiation of ESCs are major obstacles against clinical uses of ESC-based therapies. A cohort of cell surface bound and soluble factors, interactions of ESCs with their neighboring cells and extracellular matrix proteins, and various epigenetic factors may act synergistically to drive differentiation of stem cells. While most of current research is centered on functionalizing specific biomolecules on scaffolds and tuning the matrix stiffness, or altering media compositions to gain a better control over the neural differentiation of stem cells, the role of niche-mediated factors is less understood. In this study, we showed that intrinsic niche parameters such as stem cell colony size and interspacing between the two colonies can significantly impact the differentiation effic (open full item for complete abstract)

Committee: Hossein Tavana (Advisor); Marnie Saunders (Committee Member); Nic Leipzig (Committee Member); Yang Yun (Committee Member); Sailaja Paruchuri (Committee Member) Subjects: Biomedical Engineering; Biomedical Research; Engineering; Neurobiology

Doctor of Philosophy (PhD), Ohio University, 2018, Electrical Engineering & Computer Science (Engineering and Technology)

Changes in the morphology of a skin lesion is indicative of melanoma, a deadly type of skin cancer. This dissertation proposes a temporal analysis method to monitor the vascularity, pigmentation, size and other critical morphological attributes of the lesion. Digital images of a skin lesion acquired during follow-up imaging sessions are input to the proposed system. The images are pre-processed to normalize variations introduced over time. The vascularity is modelled as the skin images' red channel information and its changes by the Kullback-Leibler (KL) divergence of the probability density function approximation of histograms. The pigmentation is quantified as textural energy, changes in the energy and pigment coverage in the lesion. An optical flow field and divergence measure indicates the magnitude and direction of global changes in the lesion. Sub-surface change is predicted based on the surface skin lesion image with a novel approach. Changes in key morphological features such as lesions' shape, color, texture, size, and border regularity are computed. Future trends of the skin lesions features are estimated by an auto-regressive predictor. Finally, the features extracted using deep convolutional neural networks and the hand-crafted lesion features are compared with classification metrics. An accuracy of 80.5%, specificity of 98.14%, sensitivity of 76.9% with a deep learning neural network is achieved. Experimental results show the potential of the proposed method to monitor a skin lesion in real-time during routine skin exams.

Committee: Mehmet Celenk Ph.D. (Advisor); Savas Kaya Ph.D. (Committee Member); Jundong Liu Ph.D. (Committee Member); Razvan Bunescu Ph.D. (Committee Member); Xiaoping Shen Ph.D. (Committee Member); Sergio Lopez-Permouth Ph.D. (Committee Member) Subjects: Computer Science; Electrical Engineering; Medical Imaging; Oncology

Doctor of Philosophy (Ph.D.), University of Dayton, 2018, Engineering

All remote sensing systems and sensors contend with unwanted signals, broadly categorized as distortions, that contribute error to the ideally sensed image. Great effort is dedicated by system designers and end-users toward characterizing and mitigating distortions, which facilitates the ability to confidently exploit the sensed data. Distortions to the ideal image can be categorized into two classes: observed and constructed. Observed distortion is the class of distortions that contains all unwanted errors within an image due to uncontrolled processes that impede an ideal reconstruction of the scene such as, but not limited to, illumination changes, perspective changes, seasonal changes, atmospheric and meteorological effects, occlusions, blur, lens distortion, and various noise sources such as shot noise, thermal noise, and fixed-pattern noise. Because observed distortions occur before or during the digitization of the scene, only estimates of the undistorted ideal image can be made. In contrast to observed distortion, constructed distortion is a class of distortion that is intentionally introduced into the scene, possibly as a byproduct of other processing. If the data needs to be transmitted from a sensing platform to another location, or if the data needs to be more efficiently stored, compression is often used. If lossy algorithms are selected for compression, the data incurs distortion. However, since the undistorted image is known and the distortion is added to the image, compression distortion is categorized as constructed distortion. Despite being able to estimate the significance of compression distortion relative to the undistorted image, compression distortion is often uncharacterized and uncontrolled due to limited knowledge on its relationship to image exploitation algorithms. This gap in knowledge is the focus of this research. The presented work is developed with the intent of bridging the fields of lossy image compression with automated image (open full item for complete abstract)

Committee: Eric Balster Ph.D. (Committee Chair); Frank Scarpino Ph.D. (Committee Member); Keigo Hirakawa Ph.D. (Committee Member); Clark Taylor Ph.D. (Committee Member) Subjects: Remote Sensing

PhD, University of Cincinnati, 2017, Arts and Sciences: Geography

The Earth stands on the precipice of the sixth mass extinction. This extinction risk has triggered a growing crisis and urgent need to save the world's biodiversity. Considering the accelerated rates of biodiversity loss and extinction, we need simple but efficient methods to quickly identify threatened areas. This dissertation research was undertaken with this in mind—to benefit the conservation community, either through the delivery of biogeographic methods or information to further the restoration or maintenance of biodiversity. As a primary goal, this dissertation endeavored to fill those research gaps and offer some simpler and more effective useful and usable geospatial techniques for biodiversity conservation analyses. Secondary goals of the research were (1) to contribute to specific conservation programs for critically endangered species, (2) to inform about the status of habitat, and (3) to address top conservation research priorities. While not a specific objective, the research outcomes may influence public policy. This three-article dissertation introduces two novel techniques: (1) development of a habitat suitability model in ArcGIS using kernel density estimation and a mortality-risk weighting factor on road density, the delimiting variable; and (2) a rapid hybrid change detection technique using ENVI's SPEAR Vegetation Delineation tool or classifying live green vegetation and ArcGIS to compare and quantify changes in time. For the latter, two studies incorporated the change detection technique. The pilot study performed the change detection with color-infrared aerial photography, while the follow-up investigation tested the feasibility of the method to handle high resolution multi-sensor data, given the difficulty obtaining data from the same or similar sensors. These studies represent the first of their kind. This dissertation research provides widely applicable, practical, and employable geospatial models to perform habitat assessment (open full item for complete abstract)

Committee: Tak Yung Tong Ph.D. (Committee Chair); Richard Beck Ph.D. (Committee Member); Theresa Culley Ph.D. (Committee Member); Nicholas Dunning Ph.D. (Committee Member); Hongxing Liu Ph.D. (Committee Member) Subjects: Geography

Doctor of Philosophy, The Ohio State University, 2017, Electrical and Computer Engineering

Wireless sensor networks have matured over the last several years from popular research and development platforms to commercially-available sensors and systems. In many applications, wireless sensor networks have size, weight, power, and cost limitations. These constraints directly affect the ability of sensor nodes to adequately process and reliably communicate information within the sensor network. This dissertation examines aspects of distributed detection and estimation over a sensor network while considering limitations inherent in wireless networks. First, we consider the problem of distributed detection from a large network of sensors and introduce a realistic network model. Sensor nodes make individual decisions from their local observation and then communicate these decisions through a shared and imperfect communications channel to a central decision node. The key difference from previous research is the network model allows the decision rule to leverage errors in the channel to improve detection performance. We derive analytical expressions that characterize the detection performance of the system with respect to both sensor density and communications delay. We show that the detection performance improves with network density when sensor nodes are appropriately censored and desensitized, despite increasing message collisions. In addition, we show that detection performance using the protocol model, with imperfect communications, rapidly converges to the perfect communications case as the number of communication slots increase. Second, we study the problem of distributed quickest change detection from a network of sensors. Similar to the first part, sensor nodes communicate information to a central decision node, but in this part the central node continues to collect information from the sensor nodes until a detection is declared. We consider a minimax-type distributed quickest change detection solution that minimizes detection delay for a d (open full item for complete abstract)

Committee: Randolph Moses (Advisor); Emre Ertin (Advisor); Eylem Ekici (Committee Member) Subjects: Electrical Engineering; Statistics

Master of Science in Computer Engineering, University of Dayton, 2016, Electrical and Computer Engineering

Many applications in infrastructure planning and maintenance are currently aided by the collection of aerial image data and manual examination by human analysts. The increasing availability and quality of this image data presents an opportunity for computer vision and machine learning techniques to aid in infrastructure planning and maintenance. Due to the immense effort required for human analysts to view and organize the data, there is great demand for computer automation of these tasks. A strategy for detecting changes in known building regions in multitemporal visible and near-infrared imagery based on a linear combination of independent features and a difference of Gaussian based classification approach is proposed. Initial building candidates are discovered using a linear combination of features representing vegetation intensity, image texture, shadow intensity and distance from known road areas. The resulting building candidates are classified by shape using a unique difference of Gaussians technique and a standard Support Vector Machine classifier. Building regions reported in the reference data set from the prior observation time are revisited using the same classification approach to minimize the number of false positive detections from the feature fusion strategy. The effectiveness of the proposed technique is evaluated on five wide area real-world images. Ground truths for the building regions in all five images are manually created and used to measure the accuracy of the building detection and change detection results. Detection statistics and visualized results of the proposed algorithm are presented, and it is observed that the results are promising compared to the manually created ground truth. As a possible continuation of this research, a brief discussion on parameter estimation for building change detection based on image characteristics is included.

Committee: Vijayan Asari Ph.D. (Advisor); Theus Aspiras Ph.D. (Committee Member); Russell Hardie Ph.D. (Committee Member) Subjects: Computer Engineering; Electrical Engineering

Doctor of Philosophy (Ph.D.), University of Dayton, 2016, Engineering

Change detection is a popular and challenging research field in which the changes between two images of the same location collected at different times are detected. The crux of the problem is to detect only significant changes, based on application goals, while suppressing insignificant changes. Insignificant changes may be related to a variety of items including illumination changes, natural seasonal changes, atmospheric conditions and image registration errors. This dissertation involves further investigation into ratio-based change detection. Standard ratio-based change detection methods use a single ratio along with a threshold and its reciprocal to detect changes in both tails of the ratio distribution. Ratio-based methods show unique ability among change detection algorithms from published literature to detect challenging changes. However, the ratio-based methods in literature also exhibit problems detecting certain types of changes and suffer from high false alarm rates. A multi-ratio fusion framework for robust change detection, based on ratio-based change detection, is proposed and tested in this dissertation. A method called dual ratio (DR) change detection is developed featuring two ratios coupled with adaptive thresholds to maximize detected changes and minimize false alarms. The use of two ratios is shown to outperform the single ratio case when the means of the image pairs are not equal. A multi-ratio (MR) change detection method is developed building upon the DR method by including negative imagery to produce four ratios with adaptive thresholds. Inclusion of negative imagery is shown to improve detection sensitivity and to boost detection performance in certain target and background cases. A multi-ratio fusion (MRF) change detection technique further expands the algorithmic concepts in DR and MR by fusing together the ratio outputs to maximize detections and minimize false alarms. In the fusion algorithm, detections must be verified by tw (open full item for complete abstract)

Committee: Eric Balster Ph.D. (Committee Chair); Vijay Asari Ph.D. (Committee Member); Juan Vasquez Ph.D. (Committee Member); Frank Scarpino Ph.D. (Committee Member) Subjects: Electrical Engineering

Doctor of Philosophy (PhD), Ohio University, 2015, Experimental Psychology (Arts and Sciences)

Longitudinal data collection is becoming increasingly common with the increased use of internet-based/technologically-based methods for data capture. In fields as diverse as healthcare, engineering, fisheries management, political science, economics, and psychology, often analyses are conducted to determine if some change to the pattern of incoming data has occurred. If a change has occurred analysis should make that determination as quickly as possible. A data-pattern change is critical information, as it may indicate a change in the health status of patients, changing political attitudes, or, as in the case of the proposed study, changes to the safety profile of consumer products. The methods to analyze these longitudinal databases for indicators of change are as varied as the fields collecting the data. To date, no single study has examined the varied methodologies to determine the relative accuracy of the methods and no study has attempted to determine the relative duration over which accurate change determinations are made. This study examined the performance of these methodologies across three sets of simulated data as well as a single, large-scale safety database for a major consumer healthcare company. The simulated data is comprised of random noise data streams and data streams with actual changes in data pattern (signals). The three simulated data sets differ by the strength of the signal. The consumer safety database is comprised of call center data (n>725,000 records) from consumers who call to report a side effect (adverse event) while taking a company product. Healthcare professionals flag products identified as having a confirmed safety signal. Analyses were conducted retrospectively to determine if this change in safety status could have been detected by the statistical methods examined in this study for 30 days prior to the date of the confirmed signal. For each of the three simulated data sets and the actual product safety database, m (open full item for complete abstract)

Committee: Bruce Carlson PhD (Committee Chair) Subjects: Quantitative Psychology

Doctor of Philosophy, The Ohio State University, 2015, Civil Engineering

Landslides are natural disasters that cause environmental and infrastructure damage worldwide. In order to reduce future risk posed by them, effective detection and monitoring methods are needed. Landslide susceptibility and hazard mapping is a method for identifying areas suspect to landslide activity. This task is typically performed in a manual, semi-automatic or automatic form, or a combination of these, and can be accomplished using different sensors and techniques. As landslide hazards continue to impact our environment and impede the lives of many, it is imperative to improve the tools and methods of effective and reliable detecting of such events. Recent developments in remote sensing have significantly improved topographic mapping capabilities, resulting in higher spatial resolution and more accurate surface representations. Dense 3D point clouds can be directly obtained by airborne Light Detection and Ranging (LiDAR) or created photogrammetrically, allowing for better exploitation of surface morphology. The potential of extracting spatial features typical to landslides, especially small scale failures, provides a unique opportunity to advance landslide detection, modeling, and prediction process. This dissertation topic selection was motivated by three primary reasons. First, 3D data structures, including data representation, surface morphology, feature extraction, spatial indexing, and classification, in particular, shape-based grouping, based on LiDAR data offer a unique opportunity for many 3D modeling applications. Second, massive 3D data, such as point clouds or surfaces obtained by the state-of-the-art remote sensing technologies, have not been fully exploited for landslide detection and monitoring. Third, unprecedented advances in LiDAR technology and availability to the broader mapping community should be explored at the appropriate level to assess the current and future advantages and limitations of LiDAR-based detection and modeling of land (open full item for complete abstract)

Committee: Dorota Grejner-Brzezinska (Advisor); Charles Toth (Advisor); Tien Wu (Committee Member) Subjects: Civil Engineering

Doctor of Philosophy (Ph.D.), University of Dayton, 2015, Electrical Engineering

We present a 3D change detection technique designed to support various wide-area-surveillance (WAS) applications in changing environmental conditions. The novelty of the work lies in our approach of creating an illumination invariant system tasked with detecting changes in a scene. Previous efforts have focused on image enhancement techniques that manipulate the intensity values of the image to create a more controlled and unnatural illumination. Since most applications require detecting changes in a scene irrespective of the time of day, (lighting conditions or weather conditions present at the time of the frame capture), image enhancement algorithms fail to suppress the illumination differences enough for Background Model (BM) subtraction to be effective. A more effective change detection technique utilizes the 3D scene reconstruction capabilities of structure from motion to create a 3D background model of the environment. By rotating and computing the projectile of the 3D model, previous work has been shown to effectively eliminate the background by subtracting the newly captured dataset from the BM projectile leaving only the changes within the scene. Although previous techniques have proven to work in some cases, these techniques fail when the illumination significantly changes between the capture of the datasets. Our approach completely eliminates the illumination challenges from the change detection problem. The algorithm is based on our previous work in which we have shown a capability to reconstruct a surrounding environment in near real-time speeds. The algorithm, namely Dense Point-Cloud Representation (DPR), allows for a 3D reconstruction of a scene using only a single moving camera. Utilizing video frames captured at different points in time allows us to determine the relative depths in a scene. The reconstruction process resulting in a point-cloud is computed based on SURF feature matching and depth triangulation analysis. We utilized optical flow fea (open full item for complete abstract)

Committee: Vijayan Asari Ph.D. (Committee Chair); Raul Ordonez Ph.D. (Committee Member); Eric Balster Ph.D. (Committee Member); Juan Vasquez Ph.D. (Committee Member) Subjects: Electrical Engineering