Signal Processing Commons^™

An Adaptive Approach To Gibbs’ Phenomenon, Jannatul Ferdous Chhoa

Master's Theses

Gibbs’ Phenomenon, an unusual behavior of functions with sharp jumps, is encountered while applying the Fourier Transform on them. The resulting reconstructions have high frequency oscillations near the jumps making the reconstructions far from being accurate. To get rid of the unwanted oscillations, we used the Lanczos sigma factor to adjust the Fourier series and we came across three cases. Out of the three, two of them failed to give us the right reconstructions because either it was removing the oscillations partially but not entirely or it was completely removing them but smoothing out the jumps a little too much. …

Cup-Net: Compressed Ultrafast Photography Using Convolutional Neural Networks, Matthew Parker

ENGS 88 Honors Thesis (AB Students)

Compressed ultrafast photography (CUP) is a cutting-edge imaging technique that uses a variation of the traditional streak camera to obtain video at 100 billion frames per second with a single exposure. In order to achieve this level of temporal detail, CUP leverages compressed sensing (CS). Compressed sensing theory states that a compressed representation of an image can be directly acquired using a non-adaptive measurement matrix so long as the encoding matrix follows certain properties such as restrictive isometry and incoherence. This compressed representation of the original scene can later be reconstructed back into the original form. CUP applies CS by …

Autoassociative-Heteroassociative Neural Network, Claudia V. Kropas-Hughes, Steven K. Rogers, Mark E. Oxley, Matthew Kabrisky

AFIT Patents

An efficient neural network computing technique capable of synthesizing two sets of output signal data from a single input signal data set. The method and device of the invention involves a unique integration of autoassociative and heteroassociative neural network mappings, the autoassociative neural network mapping enabling a quality metric for assessing the generalization or prediction accuracy of the heteroassociative neural network mapping.

Heel Down And Toe-Off Time Measured With Ultrasonic Doppler System And Force Plate Sensor, Sabin Timsina

Honors Theses

Collie Box is a medical device that measures the gait parameters of the person walk- ing in front of it. This device uses the Ultrasonic Doppler system to extract the heel-contact and toe-off times of a person walking within the range of 2-10 meters. These times are used to determine the leg’s swing phase and double stance times. The ultrasonic transducer of 10mm diameter is driven at 40kHz. At the time of the heel-contact and toe-off, foot velocity is zero while the torso part of the human body is still in motion. The wide directivity of 10mm diameter ultrasonic transducer …

An Fpga-Based Hardware Accelerator For The Digital Image Correlation Engine, Keaten Stokke

Graduate Theses and Dissertations

The work presented in this thesis was aimed at the development of a hardware accelerator for the Digital Image Correlation engine (DICe) and compare two methods of data access, USB and Ethernet. The original DICe software package was created by Sandia National Laboratories and is written in C++. The software runs on any typical workstation PC and performs image correlation on available frame data produced by a camera. When DICe is introduced to a high volume of frames, the correlation time is on the order of days. The time to process and analyze data with DICe becomes a concern when …

Development Of Fully Balanced Ssfp And Computer Vision Applications For Mri-Assisted Radiosurgery (Mars), Jeremiah Sanders

Dissertations & Theses (Open Access)

Prostate cancer is the second most common cancer in men and the second-leading cause of cancer death in men. Brachytherapy is a highly effective treatment option for prostate cancer, and is the most cost-effective initial treatment among all other therapeutic options for low to intermediate risk patients of prostate cancer. In low-dose-rate (LDR) brachytherapy, verifying the location of the radioactive seeds within the prostate and in relation to critical normal structures after seed implantation is essential to ensuring positive treatment outcomes.

One current gap in knowledge is how to simultaneously image the prostate, surrounding anatomy, and radioactive seeds within the …

Early Warning Solar Storm Prediction, Ian D. Lumsden, Marvin Joshi, Matthew Smalley, Aiden Rutter, Ben Klein

Chancellor’s Honors Program Projects

No abstract provided.

Learning Set Representations For Lwir In-Scene Atmospheric Compensation, Nicholas M. Westing [*], Kevin C. Gross, Brett J. Borghetti, Jacob A. Martin, Joseph Meola

Faculty Publications

Atmospheric compensation of long-wave infrared (LWIR) hyperspectral imagery is investigated in this article using set representations learned by a neural network. This approach relies on synthetic at-sensor radiance data derived from collected radiosondes and a diverse database of measured emissivity spectra sampled at a range of surface temperatures. The network loss function relies on LWIR radiative transfer equations to update model parameters. Atmospheric predictions are made on a set of diverse pixels extracted from the scene, without knowledge of blackbody pixels or pixel temperatures. The network architecture utilizes permutation-invariant layers to predict a set representation, similar to the work performed …

Deep Cellular Recurrent Neural Architecture For Efficient Multidimensional Time-Series Data Processing, Lasitha S. Vidyaratne

Electrical & Computer Engineering Theses & Dissertations

Efficient processing of time series data is a fundamental yet challenging problem in pattern recognition. Though recent developments in machine learning and deep learning have enabled remarkable improvements in processing large scale datasets in many application domains, most are designed and regulated to handle inputs that are static in time. Many real-world data, such as in biomedical, surveillance and security, financial, manufacturing and engineering applications, are rarely static in time, and demand models able to recognize patterns in both space and time. Current machine learning (ML) and deep learning (DL) models adapted for time series processing tend to grow in …

Development Of Gaussian Learning Algorithms For Early Detection Of Alzheimer's Disease, Chen Fang

FIU Electronic Theses and Dissertations

Alzheimer’s disease (AD) is the most common form of dementia affecting 10% of the population over the age of 65 and the growing costs in managing AD are estimated to be $259 billion, according to data reported in the 2017 by the Alzheimer's Association. Moreover, with cognitive decline, daily life of the affected persons and their families are severely impacted. Taking advantage of the diagnosis of AD and its prodromal stage of mild cognitive impairment (MCI), an early treatment may help patients preserve the quality of life and slow the progression of the disease, even though the underlying disease cannot …

Object Detection With Deep Learning To Accelerate Pose Estimation For Automated Aerial Refueling, Andrew T. Lee

Theses and Dissertations

Remotely piloted aircraft (RPAs) cannot currently refuel during flight because the latency between the pilot and the aircraft is too great to safely perform aerial refueling maneuvers. However, an AAR system removes this limitation by allowing the tanker to directly control the RP A. The tanker quickly finding the relative position and orientation (pose) of the approaching aircraft is the first step to create an AAR system. Previous work at AFIT demonstrates that stereo camera systems provide robust pose estimation capability. This thesis first extends that work by examining the effects of the cameras' resolution on the quality of pose …

Extracting Range Data From Images Using Focus Error, Erik M. Madden

Theses and Dissertations

Air-to-air refueling (AAR) has become a staple when performing long missions with aircraft. With modern technology, however, people have begun to research how to perform this task autonomously. Automated air-to-air refueling (A3R) is this exact concept. Combining many different systems, the idea is to allow computers on the aircraft to link up via the refueling boom, refuel, and detach before resuming pilot control. This document lays out one of the systems that is needed to perform A3R, namely, the system that extracts range data. While stereo cameras perform such tasks, there is interest in finding other ways of accomplishing the …

One-Dimensional Multi-Frame Blind Deconvolution Using Astronomical Data For Spatially Separable Objects, Marc R. Brown

Theses and Dissertations

Blind deconvolution is used to complete missions to detect adversary assets in space and to defend the nation's assets. A new algorithm was developed to perform blind deconvolution for objects that are spatially separable using multiple frames of data. This new one-dimensional approach uses the expectation-maximization algorithm to blindly deconvolve spatially separable objects. This object separation reduces the size of the object matrix from an NxN matrix to two singular vectors of length N. With limited knowledge of the object and point spread function the one-dimensional algorithm successfully deconvolved the objects in both simulated and laboratory data.

Maximizing Accuracy Through Stereo Vision Camera Positioning For Automated Aerial Refueling, Kirill A. Sarantsev

Theses and Dissertations

Aerial refueling is a key component of the U.S. Air Force strategic arsenal. When two aircraft interact in an aerial refueling operation, the accuracy of relative navigation estimates are critical for the safety, accuracy and success of the mission. Automated Aerial Refueling (AAR) looks to improve the refueling process by creating a more effective system and allowing for Unmanned Aerial Vehicle(s) (UAV) support. This paper considers a cooperative aerial refueling scenario where stereo cameras are used on the tanker to direct a \boom" (a large, long structure through which the fuel will ow) into a port on the receiver aircraft. …

Simulation Of Sporadic-E Parameters Using Phase Screen Method, Daniel W. Stambovsky

Theses and Dissertations

A phase screen simulation experiment is designed and implemented to model radio occultation through sporadic-E ionospheric disturbances between a GPS transmitter operating at the L1 frequency and a second receiving satellite in low earth orbit (LEO). Simulations were made to test the linear relationship between plasma intensity and scintillation S4 index both posited (Arras and Wickert, 2018) and contended (Gooch et al., 2020) in previous literature. Results brought into question both the linear relationship and the use of S4 as a whole and an alternate metric was sought.

Use Of Lidar In Automated Aerial Refueling To Improve Stereo Vision Systems, Michael R. Crowl

Theses and Dissertations

The United States Air Force (USAF) executes five Core Missions, four of which depend on increased aircraft range. To better achieve global strike and reconnaissance, unmanned aerial vehicles (UAVs) require aerial refueling for extended missions. However, current aerial refueling capabilities are limited to manned aircraft due to technical difficulties to refuel UAVs mid-flight. The latency between a UAV operator and the UAV is too large to adequately respond for such an operation. To overcome this limitation, the USAF wants to create a capability to guide the refueling boom into the refueling receptacle. This research explores the use of light detection …

Measuring Localization Confidence For Quantifying Accuracy And Heterogeneity In Single-Molecule Super-Resolution Microscopy, Hesam Mazidi, Tianben Ding, Arye Nehorai, Matthew D. Lew

Electrical & Systems Engineering Publications and Presentations

We present a computational method, termed Wasserstein-induced flux (WIF), to robustly quantify the accuracy of individual localizations within a single-molecule localization microscopy (SMLM) dataset without ground- truth knowledge of the sample. WIF relies on the observation that accurate localizations are stable with respect to an arbitrary computational perturbation. Inspired by optimal transport theory, we measure the stability of individual localizations and develop an efficient optimization algorithm to compute WIF. We demonstrate the advantage of WIF in accurately quantifying imaging artifacts in high-density reconstruction of a tubulin network. WIF represents an advance in quantifying systematic errors with unknown and complex distributions, …

Coverage Guided Differential Adversarial Testing Of Deep Learning Systems, Jianmin Guo, Houbing Song, Yue Zhao, Yu Jiang

Publications

Deep learning is increasingly applied to safety-critical application domains such as autonomous cars and medical devices. It is of significant importance to ensure their reliability and robustness. In this paper, we propose DLFuzz, the coverage guided differential adversarial testing framework to guide deep learing systems exposing incorrect behaviors. DLFuzz keeps minutely mutating the input to maximize the neuron coverage and the prediction difference between the original input and the mutated input, without manual labeling effort or cross-referencing oracles from other systems with the same functionality. We also design multiple novel strategies for neuron selection to improve the neuron coverage. The …

Image Restoration Using Automatic Damaged Regions Detection And Machine Learning-Based Inpainting Technique, Chloe Martin-King

Computational and Data Sciences (PhD) Dissertations

In this dissertation we propose two novel image restoration schemes. The first pertains to automatic detection of damaged regions in old photographs and digital images of cracked paintings. In cases when inpainting mask generation cannot be completely automatic, our detection algorithm facilitates precise mask creation, particularly useful for images containing damage that is tedious to annotate or difficult to geometrically define. The main contribution of this dissertation is the development and utilization of a new inpainting technique, region hiding, to repair a single image by training a convolutional neural network on various transformations of that image. Region hiding is also …

Average Speech Directivity, Samuel D. Bellows, Claire M. Pincock, Jennifer K. Whiting, Timothy W. Leishman

Directivity

Speech directivity describes the angular dependence of acoustic radiation from a talker’s mouth and nostrils and diffraction about his or her body and chair (if seated). It is an essential physical aspect of communication affecting sounds and signals in acoustical environments, audio, and telecommunication systems. Because high-resolution, spherically comprehensive measurements of live, phonetically balanced speech have been unavailable in the past, the authors have undertaken research to produce and share such data for simulations of acoustical environments, optimizations of microphone placements, speech studies, and other applications. The measurements included three male and three female talkers who repeated phonetically balanced passages …

Paper-Based Flexible Electrode Using Chemically-Modified Graphene And Functionalized Multiwalled Carbon Nanotube Composites For Electrophysiological Signal Sensing, Md Faruk Hossain, Jae Sang Heo, John Nelson, Insoo Kim

Bioelectrics Publications

Flexible paper-based physiological sensor electrodes were developed using chemically-modified graphene (CG) and carboxylic-functionalized multiwalled carbon nanotube composites (f@MWCNTs). A solvothermal process with additional treatment was conducted to synthesize CG and f@MWCNTs to make CG-f@MWCNT composites. The composite was sonicated in an appropriate solvent to make a uniform suspension, and then it was drop cast on a nylon membrane in a vacuum filter. A number of batches (0%~35% f@MWCNTs) were prepared to investigate the performance of the physical characteristics. The 25% f@MWCNT-loaded composite showed the best adhesion on the paper substrate. The surface topography and chemical bonding of the proposed CG-f@MWCNT …

Demonstration Of Visible And Near Infrared Raman Spectrometers And Improved Matched Filter Model For Analysis Of Combined Raman Signals, Alexander Matthew Atkinson

Electrical & Computer Engineering Theses & Dissertations

Raman spectroscopy is a powerful analysis technique that has found applications in fields such as analytical chemistry, planetary sciences, and medical diagnostics. Recent studies have shown that analysis of Raman spectral profiles can be greatly assisted by use of computational models with achievements including high accuracy pure sample classification with imbalanced data sets and detection of ideal sample deviations for pharmaceutical quality control. The adoption of automated methods is a necessary step in streamlining the analysis process as Raman hardware becomes more advanced. Due to limits in the architectures of current machine learning based Raman classification models, transfer from pure …

A Harmless Wireless Quantum Alternative To Cell Phones Based On Quantum Noise, Florentin Smarandache, Robert Neil Boyd, Victor Christianto

Branch Mathematics and Statistics Faculty and Staff Publications

In the meantime we know that 4G and 5G technologies cause many harms to human health. Therefore, here we submit a harmless wireless quantum alternative to cell phones. It is our hope that this alternative

Digital Holography Efficiency Experiments For Tactical Applications, Douglas E. Thornton

Theses and Dissertations

Digital holography (DH) uses coherent detection and offers direct access to the complex-optical field to sense and correct image aberrations in low signal-to-noise environments, which is critical for tactical applications. The performance of DH is compared to a similar, well studied deep-turbulence wavefront sensor, the self-referencing interferometer (SRI), with known efficiency losses. Wave optics simulations with deep-turbulence conditions and noise were conducted and the results show that DH outperforms the SRI by 10's of dB due to DH's strong reference. Additionally, efficiency experiments were conducted to investigate DH system losses. The experimental results show that the mixing efficiency (37%) is …

The Challenge For Vision Of Fluctuating Real-World Illumination, David H. Foster

MODVIS Workshop

No abstract provided.

An Underground Radio Wave Propagation Prediction Model For Digital Agriculture, Abdul Salam

Faculty Publications

Underground sensing and propagation of Signals in the Soil (SitS) medium is an electromagnetic issue. The path loss prediction with higher accuracy is an open research subject in digital agriculture monitoring applications for sensing and communications. The statistical data are predominantly derived from site-specific empirical measurements, which is considered an impediment to universal application. Nevertheless, in the existing literature, statistical approaches have been applied to the SitS channel modeling, where impulse response analysis and the Friis open space transmission formula are employed as the channel modeling tool in different soil types under varying soil moisture conditions at diverse communication distances …

Generating Spectra Using Pca-Based Spectral Mixture Models, Joseph S. Makarewicz, Heather D. Makarewicz

Scholar Week 2016 - present

PCA-based spectra mixture models have been created for several laboratory mixture data sets. This presentation provides examples of spectra that were generated using PCA-based spectra mixture models.

Insar Simulations For Swot And Dual Frequency Processing For Topographic Measurements, Gerard Masalias Huguet

Masters Theses

In Earth remote sensing precise characterization of the backscatter coefficient is important to extract valuable information about the observed target. A system that eliminates platform motion during near-nadir airborne observations is presented in this thesis, showing an improvement on the accuracy of measurements for a Ka- band scatterometer previously developed at Microwave Remote Sensing Laboratory (MIRSL). These very same results are used to simulate the reflectivity of such targets as seen from a spaceborne radar and estimate height errors based on mission-specific geometry. Finally, data collected from a dual-frequency airborne interferometer com- prised by the Ka-band system and an S-band …

Improving Detection Of Dim Targets: Optimization Of A Moment-Based Detection Algorithm, Shannon R. Young

Theses and Dissertations

Wide area motion imagery (WAMI) sensor technology is advancing rapidly. Increases in frame rates and detector array sizes have led to a dramatic increase in the volume of data that can be acquired. Without a corresponding increase in analytical manpower, much of these data remain underutilized. This creates a need for fast, automated, and robust methods for detecting dim, moving signals of interest. Current approaches fall into two categories: detect-before-track (DBT) and track-before-detect (TBD) methods. The DBT methods use thresholding to reduce the quantity of data to be processed, making real time implementation practical but at the cost of the …

End-To-End Deep Learning Systems For Scene Understanding, Path Planning And Navigation In Fire Fighter Teams, Manish Bhattarai

Shared Knowledge Conference

Firefighting is a dynamic activity with many operations occurring simultaneously. Maintaining situational awareness, defined as knowledge of current conditions and activities at the scene, are critical to accurate decision making. Firefighters often carry various sensors in their personal equipment, namely thermal cameras, gas sensors, and microphones. Improved data processing techniques can mine this data more effectively and be used to improve situational awareness at all times thereby improving real-time decision making and minimizing errors in judgment induced by environmental conditions and anxiety levels. This objective of this research employs state of the art Machine Learning (ML) techniques to create an …