Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “Image Analysis”

Search indexed NASA NTRS and DOE OSTI research on propulsion, heat transfer, battery materials and energy systems. Follow report and document links to the original sources.

Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.

At least 1,333 records · Page 74
All sourcesNASA NTRSDOE OSTI
NASA NTRS2015

Assessment of the 2013-2014 Drought on Ecosystems of the California Central Coast

Calendar year 2013 was the driest on record in California, with a total of just 30 percent of average statewide precipitation. The objective of this present study was to assess the impacts of the historic 2013-2014 drought on ecosystems of the Central Coast region using a combination of satellite image analysis and in situ measurements of soil moisture. According to differences in Landsat NDWI and NDVI between May of 2010 and 2013, the geographic areas within the study region that were most severely impacted by the 2013 drought were the inland Carmel Valley in northern Monterey County, and the coast zones around San Simeon Point and Cambria in northern San Luis Obispo County. For more detailed examination of drought impacts, the entire study region was separated into the three predominate vegetation types (grasslands, shrublands, and forests) to examine changes in Landsat NDWI and NDVI in the context of differing plant community response to severe drought.

Normalized Differences Vegetation Index (NDVI)↗
NASA NTRS2017

On-Orbit Measurement of the Focal Length of the SNPP VIIRS Instrument

The Visible Infrared Imaging Radiometer Suite (VIIRS) instrument is a whiskbroom system with 22 spectral bands split between 16 moderate resolution bands (M-bands), five imagery resolution bands (I-bands) and a panchromatic day-night band. Latitude and Longitude geolocation data are generated for each pixel at the M-band, I-band and day-night band spatial resolutions based upon various instrument parameters including focal length. In this study we measure the focal length of the VIIRS instrument from on-orbit data. This is achieved by simulating VIIRS band I2 using Landsat 8 OLI band 5 utilizing the VIIRS instrument system point spread function (PSF) and geolocation data generated with varying values of focal length. The focal length value that produces the highest spatial correlation between the original and simulated VIIRS data is taken to be the measured instrument focal length.

Image analysis↗
NASA NTRS2019

Exploration of Failure and Potential Damage Markers in Ti-6Al-4V

To improve prognostics for predicting component life, coupon tests were conducted on Ti-6Al-4V to investigate the development of damage. Various test types were employed (tension, creep, stress relaxation, and fatigue) under a wide range of temperatures and loading rates to engage various amounts of time-dependent and damage behavior. Both stiffness degradation and Poisson’s ratio were monitored to see if they were useful in indicating the initiation and accumulation of damage and subsequently the onset of failure. Stiffness exhibited large decreases with increasing strain when calculated using engineering stress and strain. However, minimal stiffness change was observed when using the true-stress and true- strain values. Detailed microscopy and image analysis were performed to document the actual physical damage that occurred within the various specimens. Two types of damage were observed: shallow surface cracks and internal pores. The severity of damage was greater in the region of localized necking and diminished further away from the neck. The area fraction of pores was higher in the necked region, yet even there it did not exceed more than a few percent and, in most cases, was significantly lower. It was observed that conditions that induced more time-dependent deformation (e.g., slower strain rate testing) resulted in a larger number of pores and larger pore sizes. The pores were always found to occur in the softer β-phase. The severity of observed damage was inconsistent with the large drop in engineering stiffness, but was more aligned with stiffness changes from the true-stress and -strain values. Poisson’s ratio was monitored on every test and showed promise as a damage indicator. It was found to be more sensitive than either the ultimate tensile strength or the inception of tertiary creep. However, calculation of Poisson’s ratio was found to be more involved than expected because of anisotropy and the presence of strain gradients along the specimen at large applied strains. First, the material had a moderate texture and resulted in anisotropy of the strains. Second, the evolution of Poisson’s ratio as a function of axial strain showed strange and unexpected behaviors, which necessitated more extensive analyses of its development. This led to an investigation on the use of small versus large strain approaches and the importance of making consistent measurements; that is, axial and transverse strain measurements over the same volume of material. Additionally, finite element analysis of a tensile sample indicated that non- uniform strain occurred in the sample prior to attaining the maximum strength. This, along with the location at which transverse strain was measured, had a significant influence on the resulting values of Poisson’s ratio. Based on this, lessons learned are offered for future testing.

Ti-6Al-4V↗
NASA NTRSDate not supplied

Virtual Diagnostics Interface (ViDI) for Planning Laser-Based Wind Tunnel Experiments

Virtual Diagnostics Interface (ViDI), originally introduced by Richard Schwartz in 2004 and described again by Alderfer, et. al., in 2007, is an interactive 3D software tool developed at NASA Langley Research Center to serve three main purposes: assist in pre-test planning, enhance post-test image analysis and visualization, and serve as a central location for data storage and distribution.1,2 ViDI is composed of custom-developed software and Autodesk® 3ds Max® as it incorporates Computer Aided Design (CAD) models with the results obtained from optical diagnostics.3 This software tool has proven useful in the field of optical diagnostics where pre-test planning of beam paths, optical elements, and test articles must be carried out to avoid interference, ensure spatial resolution is sufficient, and the camera Field of View (FoV) is adequate for the field of interest. Typically, practical considerations such as the distance a camera is from the region of interest, the angles at which the camera is oriented and issues such as the size of the camera and its sensor and size and focal length lens as well as the path of the laser need to be taken into account to predict a successful measurement. An example application of ViDI is shown in Fig.1 where an experiment to investigate flow separation over the wing of the High Lift Common Research Model (HL-CRM) is currently being planned in the National Transonic Facility (NTF) at NASA Langley Research Center.

virtual diagnostics interface↗
NASA NTRS2020

A Multi-Modal Approach for Monitoring Changes in Agriculture in the Mekong River Delta

Smallholder farms in South East Asia are characterized by small irregular field patterns, dense cloud cover and haze which limits our ability to observe changes in agriculture land-use. Very-high resolution (VHR, < 5 m) Earth observing (EO) satellites have grown into constellations with global repeat coverage that can support multiple observations within a growing season. Here we present a multi-modal approach to map and monitor changes in agriculture at the field scale in the Mekong River Delta with VHR optical imagery from Worldview-1, 2 and 3, and synthetic aperture radar (SAR) from Sentinel-1. Our approach leverages the benefits of SAR and VHR imagery with geographic object-based image analysis (GEOBIA) in a high-end computing environment to rapidly map the distribution and size of agriculture fields that are not observable at moderate resolution (> 30 m) in regions with dense persistent cloud cover and haze from biomass burning.

Christopher S R Neigh↗
NASA NTRSDate not supplied

Open-source Techniques for Automated Landslide Inventory Generation for Rapid Response

Manual mapping is the most used method for generating landslide inventories. For rapid response scenario this method becomes tedious and time consuming. The Landslide team at NASA Goddard Space Flight Center has been developing open-source landslide mapping systems for rapid generation of landslide inventories. We have developed a Python-based landslide mapping framework known as the Semi-Automatic Landslide Detection (SALaD) system that uses Object-based Image Analysis and machine learning. For production of event-based inventories, SALaD was modified to include a change detection module (SALaD-CD). Utilizing high-resolution imagery form from Planet and Maxar, we have generated multiple rapid response landslide inventories that have been used by emergency responders on the ground, the NASA Disasters program, and academia. Currently, we are exploiting deep learning frameworks for landslide mapping. We are interested to learn about efficient way to harmonize multi-sensor data for creating a long-term record of landslides, training strategies and ongoing deep learning-based efforts for natural hazard characterization within NASA and UMD.

Pukar Amatya↗
NASA NTRSDate not supplied

Astrometric Reduction of Phoebe Using a Digital Shape Model

The Cassini Imaging Science Subsystem (ISS) has provided the most spectacular images of the Saturnian system [1] and these observations constitute a fundamental dataset of high accuracy and exceptionally long-time span for the purpose of astrometry [2]. Several elaborate astrometric efforts using Cassini ISS data have demonstrated their use for the development and maintenance of the Saturnian satellite ephemerides and have been critical in improving our understanding of their secular orbital evolution and interior properties[3].Astrometric efforts using images with well-resolved bodies often employ limb-based methods for center finding; this is proven to be robust for satellites with a uniform ellipsoidal shape. However, systematic biases are of concern for bodies that deviate considerably from tri-axial ellipsoids(e.g., moons with irregular shapes and/or extensive features such as craters, ridges, slumps and grooves). To help understand such biases, we used the Goddard Image Analysis and Navigation Tool (GIANT) [4,5]to improve the astrometric reduction of these moons. Here we apply the method to Phoebe (Saturn IX).

V Viswanathan↗
NASA NTRS2021

High Energy Particle Excitations as a Measure of Electronic Crosstalk in MODIS and VIIRS VIS/NIR Bands

The MODIS instruments aboard the Terra and Aqua satellites and the VIIRS instruments aboard the SNPP and NOAA-20 satellites each contain several arrays of Si detectors that measure Earth-reflected radiance in the visible and near-infrared spectral range. Even in the absence of incident light, the Si detectors are occasionally excited by high energy charged particles that pass through the spacecraft. These particle radiation events are, fortunately, infrequent enough that they do not lead to significant degradation of the detectors and they do not have a significant impact on the Earth scene radiance images. On the other hand, they are frequent enough that the cumulative data from many years on orbit may provide valuable diagnostic information about the sensors. In this paper, we provide some basic statistics on the frequency and magnitude of the particle excitation events for MODIS and VIIRS and explore the usefulness of this data as a measure of electronic crosstalk. Large amounts of crosstalk can degrade the quality of the Earth images, so it is crucial to have methods to characterize and correct for it on-orbit, which has previously been done for MODIS using lunar image analysis. The particle excitations can manifest as single-pixel spikes in the otherwise dark space view background, which may be an ideal source for evaluating crosstalk. We derive crosstalk coefficients between the NIR band detectors of Terra MODIS, and compare them to coefficients previously derived from lunar observations. The same approach is applied to SNPP VIIRS, which does not show any significant electronic crosstalk. While the HgCdTe detectors used in the MODIS and VIIRS infrared bands can also be excited by particle radiation, the magnitudes of the excitations are much smaller compared to the Si detectors and in general are not large enough to be useful for examining crosstalk.

MODIS↗
NASA NTRSDate not supplied

A Comparison of Bearing Measurements to Surface Features Generated Using Stereophotoclinometry and Surface Feature Navigation Techniques

The Origins Spectral Interpretation Resource Identification Security Regolith Explorer (OSIRIS-REx) mission to the asteroid Bennu completed successful two-and-a-half year proximity operations in May 2021. The mission comprehensively mapped Bennu at unprecedented detail and collected a sample of Bennu’s surface to return to Earth. Throughout proximity operations, the OSIRIS-REx navigation team used the maps made of Bennu’s surface to navigate in the Bennu environment with high accuracy through the use of precise and accurate optical navigation data, radiometric data, and force modelling. The primary type of optical navigation measurements extracted from the images captured by OSIRIS-REx (particularly after first entering orbit around Bennu) were observations of known features on Bennu’s surface. Two related but different techniques/tools were used to extract these observations from the images: the Goddard Image Analysis and Navigation Tool Surface Feature Navigation (GIANT SFN) and Stereophotoclinometry (SPC) Autoregister. In this paper we compare the differences between the observables extracted using GIANT SFN and SPC Autoregister, explain the differences, and discuss where each technique is best suited.

Andrew Liounis↗
NASA NTRS2021

Search for Lensing Signatures in the Gravitational-Wave Observations from the First Half of LIGO–Virgo's Third Observing Run

We search for signatures of gravitational lensing in the gravitational-wave signals from compact binary coalescences detected by Advanced Laser Interferometer Gravitational-wave Observatory (LIGO) and Advanced Virgo during O3a, the first half of their third observing run. We study: (1) the expected rate of lensing at current detector sensitivity and the implications of a non-observation of strong lensing or a stochastic gravitational-wave background on the merger-rate density at high redshift; (2) how the interpretation of individual high-mass events would change if they were found to be lensed; (3) the possibility of multiple images due to strong lensing by galaxies or galaxy clusters; and (4) possible wave-optics effects due to point-mass microlenses. Several pairs of signals in the multiple-image analysis show similar parameters and, in this sense, are nominally consistent with the strong lensing hypothesis. However, taking into account population priors, selection effects, and the prior odds against lensing, these events do not provide sufficient evidence for lensing. Overall, we find no compelling evidence for lensing in the observed gravitational-wave signals from any of these analyses.

R. Abbott↗
NASA NTRS2022

Radiation Tolerance and Mitigation for Neuromorphic Processors

Neuromorphic processors are designed to execute Deep Neural Networks (DNNs) at very high speed using only a fraction of the electrical power needed to run a DNN on a traditional CPU or GPU. This unique capability makes Neuromorphic processors a prime candidate for space systems, where advanced computational tasks like image analysis, depth map reconstruction, or rover control need to be executed in a power-starved environment. In contrast to the growing number of applications of Neuromorphic processors in smart phones, the automotive and robotics domain, the space environment is unforgiving because of extreme temperatures and high levels of radiation. Any space system, operating beyond LEO requires computing hardware that is resilient against radiation effects. However, Neuromorphic processors have not yet been designed or tested for their radiation tolerance. In this report, we consider traditional methods of detection of radiation events and mitigation via redundancy and gauge their effectiveness on DNNs. In contrast to traditional flight software, however, neural networks represent a statistical algorithm, which might affect its resilience against radiation events. We will focus on the analysis of the tolerance of DNNs with respect to radiation events and discuss techniques to detect radiation hits using on-chip triple modular redundancy (TMR) on an Intel Loihi neuromorphic processor and to mitigate radiation damage. We describe an architecture for on-chip TMR for the Intel Loihi and present results of initial experiments.

Neural Networks↗
NASA NTRSDate not supplied

New Optical Navigation Results Using Historical MESSENGER Data

This paper describes new optical navigation (OpNav) results obtained by processing previously collected measurements from the MESSENGER mission to Mercury. This project also serves to mature the tools and capabilities of NASA Goddard Space Flight Center (GSFC) in OpNav, using the open source Goddard Image Analysis and Navigation Tool (GIANT). New navigation measurements are obtained during the Mercury flyby and orbital phases, using OpNav measurements generated by GIANT, and these measurements are compared to predictions. The results obtained provide a set of improvements to be made in navigation tools and will pave the way for future missions to navigate near terrestrial bodies using optical measurements.

Optical Navigation↗
NASA NTRSDate not supplied

Restoring the Dunes: Using Satellite Imagery to Help Restore Turtle Nesting Grounds In Puerto Rico After Hurricane Maria

The North coast of Puerto Rico sustained heavy damage from Hurricane Maria and other subsequent tropical storms. This damage included erosion of beaches and foredunes. The foredunes are a delicate ecological environment, which provides important nesting grounds for endangered turtles. Local ecologists have been working to restore these dunes with mixed results. Presented here is an evaluation of dune restoration sites, and strategies to improve the success of subsequent restoration efforts. This involves both historical satellite image analysis and field work to evaluate sediment migration pathways. Local ecologists use a technique called Biomimicry to restore dunes. This involves inserting sticks or plants into the ground, simulating the effects of natural vegetation to slow wind speeds and deposit wind-blown sand. Manually planted vegetation can follow to stabilize the trapped sand. While often successful, this technique occasionally fails to accumulate significant amounts of sand. Satellite imagery shows that one of the controlling factors is the orientation of the dune and beach to the prevailing wind. Dunes that are oriented parallel to the prevailing wind are relatively slower-growing due to sediment bypass. Satellite imagery also indicates that the fastest growing dunes are in close proximity to major sediment sources. This is complicated by large dams, which limit the amount of sediment delivered to the coast. Field observation shows that biomimicry matrices are occasionally located too close to the swash zone, where storms over-top the dunes, poisoning stabilizing vegetation with salt water. Another common observation was that many dunes were starting to form behind fallen trees, suggesting that a different type of biomimicry might be successful. Satellite and field observation suggest that strategies to protect nearshore environments and improve dune restoration biomimicry efforts include: 1) orient biomimicry matrices at a higher angle to prevailing winds. 2) Prioritize sites near river mouths, particularly those without upstream dams. 3) Test other orientations of biomimicry, such as horizontal boards to mimic fallen trees. 4) Improve sediment supply by preventing further dam development, and releasing trapped sediment.

Paul Bremner↗
NASA NTRS2023

The Life and Death of A Subglacial Lake in West Antarctica

Over the past 50 years, the discovery and initial investigation of subglacial lakes in Antarctica have highlighted the paleoglaciological information that may be recorded in sediments at their beds. In December 2018, we accessed Mercer Subglacial Lake, West Antarctica, and recovered the first in situ subglacial lake-sediment record—120 mm of finely laminated mud. We combined geophysical observations, image analysis, and quantitative stratigraphy techniques to estimate long-term mean lake sedimentation rates (SRs) between 0.49 ± 0.12 mm a –1 and 2.3 ± 0.2 mm a –1 , with a most likely SR of 0.68 ± 0.08 mm a –1 . These estimates suggest that this lake formed between 53 and 260 a before core recovery (BCR), with a most likely age of 180 ± 20 a BCR—coincident with the stagnation of the nearby Kamb Ice Stream. Our work demonstrates that interconnected subglacial lake systems are fundamentally linked to larger-scale ice dynamics and highlights that subglacial sediment archives contain powerful, century-scale records of ice history and provide a modern process-based analogue for interpreting paleo–subglacial lake facies.

subglacial lakes↗
NASA NTRS2024

Soil Salinity Level Assessment and Prediction Integrating UAV-borne Hyperspectral Imaging and Machine Learning Algorithms to Combat Desertification

In response to the ongoing global food crisis, the United Nations has identified “Zero Hunger” as one of its Sustainable Development Goals. A central contributor to the crisis is the process in which agricultural lands go through desertification. Research has shown a direct correlation between soil salinity and desertification - increased salinity levels indicate a higher risk for desertification. Furthermore, researchers have explored various techniques to map soil salinity, but these methods are oftentimes inefficient and don’t address future salinity predictions. To improve desertification monitoring, soil salinity can be observed via hyperspectral imaging on unmanned aerial vehicles (UAVs) to predict the risk of agricultural desertification using artificial intelligence (AI) and machine learning (ML) techniques. A significant gap exists in past research that applies ML and imaging techniques to soil salinity: convolutional neural networks (CNNs) and regression models are rarely leveraged together, despite the efficiency and accuracy of these models. To compensate for this gap, the proposed system leverages the use of these AI and ML models to improve soil assessment and prediction techniques. This approach involves three steps - data collection, image analysis, and future prediction. Using hyperspectral cameras on UAVs to collect the data from the region, a trained CNN model will output estimated soil salinity levels at a specific time. The estimations will then be analyzed by a regression model to assess the accuracy of future soil salinity predictions. The proposed system will identify regions at risk of desertification to help farmers mitigate agricultural loss, in turn helping alleviate the food crisis.

UAV systems↗
DOE OSTI2016

Learning Weight Uncertainty with Stochastic Gradient MCMC for Shape Classification

Learning the representation of shape cues in 2D & 3D objects for recognition is a fundamental task in computer vision. Deep neural networks (DNNs) have shown promising performance on this task. Due to the large variability of shapes, accurate recognition relies on good estimates of model uncertainty, ignored in traditional training of DNNs, typically learned via stochastic optimization. This paper leverages recent advances in stochastic gradient Markov Chain Monte Carlo (SG-MCMC) to learn weight uncertainty in DNNs. It yields principled Bayesian interpretations for the commonly used Dropout/DropConnect techniques and incorporates them into the SG-MCMC framework. Extensive experiments on 2D & 3D shape datasets and various DNN models demonstrate the superiority of the proposed approach over stochastic optimization. Our approach yields higher recognition accuracy when used in conjunction with Dropout and Batch-Normalization.

convolutional neural network↗
DOE OSTI2017

LAMP2GO

The app is a unique image analysis software and acts as a fluorescence reader for multiplexed nucleic acid amplification reactions. If the reaction is positive, a bright fluorescent signal is emitted from the solution depending on the choice of fluorophore molecule. The app reads the Red, blue and green (RGB) channels of the emitted signal and transforms it to yield the chromaticity (x and y) and luminance of the signal. This new representation of signal is far superior to the RGB system in determining the signal color and intensity.

Priye, Aashish↗
DOE OSTI2018

DeepPDF: A Deep Learning Approach to Extracting Text from PDFs

Scientific publications contain a plethora of important information, not only for researchers but also for their managers and institutions. Many researchers try to collect and extract this information in large enough quantities that it requires machine automation, but because publications were historically intended for print and not machine consumption, the digital document formats used today (primarily PDF) have created many hurdles for text extraction. Primarily, tools have relied on trying to convert PDF's to plain text documents for machine processing by reverse engineering the PDF standard. A complex process because once a PDF is created it is more closely related to an image file than a document markup language. In this paper we explore the feasibility of treating these PDF documents as images as opposed to a proprietary markup language. We believe that by using deep learning and image analysis we can create more accurate PDF to text extraction tools than those that currently exist. \\ \newline \Keywords{deep learning, text extraction, information extraction, PDF extraction, scholarly publications.

Stahl, Christopher↗