SEARCH · Search NASA

Results for “visual relationship detection”

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.

SourceResearch connectionFrom yearThrough year

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

At least 19 records

All sources NASA NTRS DOE OSTI

DOE OSTI2022

Tackling the Challenges in Scene Graph Generation With Local-to-Global Interactions

In this work, we seek new insights into the underlying challenges of the scene graph generation (SGG) task. Quantitative and qualitative analysis of the visual genome (VG) dataset implies: 1) ambiguity: even if interobject relationship contains the same object (or predicate), they may not be visually or semantically similar; 2) asymmetry: despite the nature of the relationship that embodied the direction, it was not well addressed in previous studies; and 3) higher-order contexts: leveraging the identities of certain graph elements can help generate accurate scene graphs. Motivated by the analysis, we design a novel SGG framework, Local-to-global interaction networks (LOGINs). Locally, interactions extract the essence between three instances of subject, object, and background, while baking direction awareness into the network by explicitly constraining the input order of subject and object. Globally, interactions encode the contexts between every graph component (i.e., nodes and edges). Finally, Attract and Repel loss is utilized to fine-tune the distribution of predicate embeddings. By design, our framework enables predicting the scene graph in a bottom-up manner, leveraging the possible complementariness. To quantify how much LOGIN is aware of relational direction, a new diagnostic task called Bidirectional Relationship Classification (BRC) is also proposed. Overall, experimental results demonstrate that LOGIN can successfully distinguish relational direction than existing methods (in BRC task), while showing state-of-the-art results on the VG benchmark (in SGG task).

97 MATHEMATICS AND COMPUTING↗

NASA NTRS1972

Fixation of virgin lunar surface soil

Two systems are shown to be suitable for fixing loose particulate soils with a polymer film, without visually detectable disturbance of the soil particle spatial relationships. A two-component system is described, which uses a gas monomer condensible at the soil temperature and a gas phase catalyst acting to polymerize the monomer. A one-component system using a monomer which polymerizes spontaneously on and within the top few millimeters of the soil is also considered. The two-component system employs a simpler apparatus, but it operates over a narrower temperature range (approximately -40 to -10 C). Other two-component systems were identified which may operate at soil temperatures as high as +100 C, at relatively narrow temperature ranges of approximately 30 C. The one-component system was demonstrated to operate successfully with initial soil temperatures from -70 C or lower to +150 C.

Conley, J. M.↗

NASA NTRS1971

Application of Infrared Scanners to Forest Fire Detection

The potential of using infrared scanners for the detection of forest fires is discussed. An experiment is described in which infrared and visual detection systems were used jointly to study timber fire detection. Many fires were detected visually but missed by the airborne IR system, and many fires were detected by the IR system but missed visually. Until more is learned about the relationship between heat output and smoke output from latent fires, the relative effectiveness of visual and IR systems cannot be determined. The 1970 tests indicated that IR used in combination with visual detection will result in a more efficient system than visual alone. Even with limited knowledge of the relative effectiveness of the two systems, operational use of a combined system can be used to substantially reduce total firefighting costs.

Hirsch, S. N.↗

DOE OSTI2025

ARCH: Large-scale knowledge graph via aggregated narrative codified health records analysis

Objective: Electronic health record (EHR) systems contain a wealth of clinical data stored as both codified data and free-text narrative notes (NLP). The complexity of EHR presents challenges in feature representation, information extraction, and uncertainty quantification. Here, to address these challenges, we proposed an efficient Aggregated naRrative Codified Health (ARCH) records analysis to generate a large-scale knowledge graph (KG) for a comprehensive set of EHR codified and narrative features. Methods: Using data from 12.5 million Veterans Affairs patients, ARCH first derives embedding vectors and generates similarities along with associated p-values to measure the strength of relatedness between clinical features with statistical certainty quantification. Next, ARCH performs a sparse embedding regression to remove indirect linkage between features to build a sparse KG. Finally, ARCH was validated on various clinical tasks, including detecting known relationships between entity pairs, predicting drug side effects, disease phenotyping, as well as sub-typing Alzheimer’s disease patients. Results: ARCH produces high-quality clinical embeddings and KG for over 60,000 codified and narrative EHR concepts. The KG and embeddings are visualized in the R-shiny powered web-API.3 ARCH achieved high accuracy in detecting EHR concept relationships, with AUCs of 0.926 (codified) and 0.861 (NLP) for similar EHR concepts, and 0.810 (codified) and 0.843 (NLP) for related pairs. It detected drug side effects with a 0.723 AUC, which improved to 0.826 after fine-tuning. Using both codified and NLP features, the detection power increased significantly. Compared to other methods, ARCH has superior accuracy and enhances weakly supervised phenotyping algorithms’ performance. Notably, it successfully categorized Alzheimer’s patients into two subgroups with varying mortality rates. Conclusion: The proposed ARCH algorithm generates large-scale high-quality semantic representations and knowledge graph for both codified and NLP EHR features, useful for a wide range of predictive modeling tasks.

Electronic health records↗

NASA NTRS1980

Thresholds for detection of constant rotary acceleration during vibratory rotary acceleration

The effects of vibratory angular acceleration on detection thresholds for constant angular acceleration in a dynamic flight simulator are reported in three experiments. Detection thresholds were determined for 10 pilots and four nonpilots using a random, double-staircase procedure while the subjects sat erect in a device which rotated about an earth-vertical axis. Constant angular acceleration were presented for 0.5 and 1.0 s with concurrent, vibratory angular acceleration at 1 and 5 Hz, and thresholds with no vibratory angular acceleration were established. The thresholds were obtained while the subjects observed a visual reference in the enclosed cockpit in two experiments and in total darkness in a third. The results confirmed earlier experiments showing an inverse relationship between the duration of constant angular acceleration and detection threshold and showed that the detection thresholds in darkness were higher than with a visual reference present. Two analyses of variance revealed no significant differences in thresholds across the three vibration conditions. These results indicate that vibratory angular acceleration of fairly high levels can be present in a dynamic flight simulator without masking the pilot's ability to detect either maneuver or disturbance motions.

Clark, B.↗

NASA NTRS2024

Big Cypress Water Resources: Using Earth Observations to Assess Water Quality in Big Cypress Reservation, FL

Upstream agricultural development and runoff are driving harmful algal blooms in the Big Cypress Reservation. The Seminole Tribe of Florida Environmental Resource Management Department monitors water quality in the Big Cypress Reservation, located on the north end of Big Cypress National Preserve. The Environmental Resource Management Department aims to incorporate Earth observations into its investigative approach, which consists of Geographic Information Systems and in situ water sampling. We assessed the feasibility of using Earth observations to measure harmful algal blooms over time and identify vulnerable areas. We analyzed data from multiple remote sensing platforms and sensors, including Landsat 8 Thermal Infrared Spectrometer(TIRS), Landsat 9 TIRS-2, Landsat 9 Operational Land Imager 2, Sentinel-2 MultiSpectral Instrument, and Sentinel-3 Ocean and Land Color Instrument. We determined that it was challenging to use Sentinel-3, Landsat 8, and Landsat 9 imagery for monitoring the canals within the Reservation, and that Sentinel-2 was the most capable platform for the study area. Using Sentinel-2, we created spectral indices for the detection of algal blooms. We used these indices to measure algal blooms at 8 stations across the Big Cypress Reservation. We then created time series and seasonal decompositions of in situ water sample data and indices to visualize relationships between datasets, along with correlation matrices. Our correlations were not conclusive. However, we found that the spectral indices have the potential to detect algal blooms. Lastly, we evaluated Maxar WorldView-3 imagery and found that using data from sensors with higher spatial resolutions would improve results.

Remote sensing↗

NASA NTRS1972

Evaluation of nondestructive testing techniques for the space shuttle nonmetallic thermal protection system

A program was conducted to evaluate nondestructive analysis techniques for the detection of defects in rigidized surface insulation (a candidate material for the Space Shuttle thermal protection system). Uncoated, coated, and coated and bonded samples with internal defects (voids, cracks, delaminations, density variations, and moisture content), coating defects (holes, cracks, thickness variations, and loss of adhesion), and bondline defects (voids and unbonds) were inspected by X-ray radiography, acoustic, microwave, high-frequency ultrasonic, beta backscatter, thermal, holographic, and visual techniques. The detectability of each type of defect was determined for each technique (when applicable). A possible relationship between microwave reflection measurements (or X-ray-radiography density measurements) and the tensile strength was established. A possible approach for in-process inspection using a combination of X-ray radiography, acoustic, microwave, and holographic techniques was recommended.

Tiede, D. A.↗

NASA NTRS1989

Circumstellar grain extinction properties of recently discovered post AGB stars

The circumstellar grains of two hot evolved post asymptotic giant branch (post AGB) stars, HD 89353 and HD 213985 were examined. From ultraviolet spectra, energy balance of the flux, and Kurucz models, the extinction around 2175 A was derived. With visual spectra, an attempt was made to detect 6614 A diffuse band absorption arising from the circumstellar grains so that we could examine the relationship of these features to the infrared features. For both stars, we did not detect any diffuse band absorption at 6614 A, implying the carrier of this diffuse band is not the carrier of the unidentified infrared features not of the 2175 A bump. The linear ultraviolet extinction of the carbon-rich star HD 89353 was determined to continue across the 2175 A region with no sign of the bump; for HD 213985 it was found to be the reverse: a strong, wide bump in the mid-ultraviolet. The 213985 bump was found to be positioned at 2340 A, longward of its usual position in the interstellar medium. Since HD 213985 was determined to have excess carbon, the bump probably arises from a carbonaceous grain. Thus, in view of the ultraviolet and infrared properties of the two post AGB stars, ubiquitous interstellar infrared emission features do not seem to be associated with the 2175 A bump. Instead, the infrared features seem related to the linear ultraviolet extinction component: hydrocarbon grains of radius less than 300 A are present with the linear HD 89353 extinction; amorphous anhydrous carbonaceous grains of radius less than 50 A might cause the shifted ultraviolet extinction bump of HD 213985.

Buss, Richard H., Jr.↗

DOE OSTI2023

A striking relationship between dust extinction and radio detection in DESI QSOs: evidence for a dusty blow-out phase in red QSOs

ABSTRACT We present the first eight months of data from our secondary target programme within the ongoing Dark Energy Spectroscopic Instrument (DESI) survey. Our programme uses a mid-infrared and optical colour selection to preferentially target dust-reddened quasi-stellar objects (QSOs) that would have otherwise been missed by the nominal DESI QSO selection. So far, we have obtained optical spectra for 3038 candidates, of which ∼70 per cent of the high-quality objects (those with robust redshifts) are visually confirmed to be Type 1 QSOs, consistent with the expected fraction from the main DESI QSO survey. By fitting a dust-reddened blue QSO composite to the QSO spectra, we find they are well-fitted by a normal QSO with up to AV ∼ 4 mag of line-of-sight dust extinction. Utilizing radio data from the LOFAR Two-metre Sky Survey (LoTSS) DR2, we identify a striking positive relationship between the amount of line-of-sight dust extinction towards a QSO and the radio detection fraction, that is not driven by radio-loud systems, redshift and/or luminosity effects. This demonstrates an intrinsic connection between dust reddening and the production of radio emission in QSOs, whereby the radio emission is most likely due to low-powered jets or winds/outflows causing shocks in a dusty environment. On the basis of this evidence, we suggest that red QSOs may represent a transitional ‘blow-out’ phase in the evolution of QSOs, where winds and outflows evacuate the dust and gas to reveal an unobscured blue QSO.

79 ASTRONOMY AND ASTROPHYSICS↗

NASA NTRS1962

Sequential Phenomena in Psychophysical Judgments: A Theoretical Analysis

This paper deals with an analysis of psychophysical detection experiments designed to assess the limit of a human observer's level of sensitivity. A mathematical theory or the detection process is introduced that, in contrast to previous theories, provides an analysis of the sequential effects observed in psychophysical data. Two variations of the detection task are considered: information feedback and no-information feedback. In the feedback situation the subject is given information concerning the correctness of his responses, whereas in the no-feedback situation he is not. Data from a visual detection experiment with no-information feedback, and from an auditory detection experiment with information feedback are analyzed in terms of the theory. Finally, some general results are derived concerning the relationship between performance in the feedback situation and the no-feedback situation.

Atkinson, R. C.↗

NASA NTRSDate not supplied

Satellite orbit predictor

An analog aid to determine satellite coverage of Emergency Locator Transmitters Emergency Position Indicating Radio Beacon (ELT/EPIRB) distress incidence is discussed. The satellite orbit predictor is a graphical aid for determining the relationship between the satellite orbit, antenna coverage of the spacecraft and coverage of the Local User Terminal. The predictor allows the user to quickly visualize if a selected position will probably be detected and is composed of a base map and a satellite track overlay for each satellite.A table of equator crossings for each satellite is included.

Friedman, Morton l.↗

NASA NTRSDate not supplied

Phenomena Portal: Large- Scale Visual Exploration of Atmospheric Phenomena

The Earth science community is experiencing a high influx of remote sensing data due to recent advancements in sensor technology. This enables the community to extend their research on a larger scale than ever before. Unfortunately, traditional data processing techniques do not scale well to these new, high volume data sources. State-of-the-art machine learning (ML) pipelines have been proven to overcome these burdens in various other fields but are underexploited within the physical sciences community. Moreover, ML is reliant on labeled data, which is currently sparsely available, owing to the fact that ML adoption is still in the early stages within the Earth and atmospheric science communities. To address these issues, we developed the Phenomena Portal, a visual exploration tool that uses ML to detect various atmospheric phenomena on a global scale. This allows the Earth and atmospheric science communities to view trends of occurrences of phenomena, identify potential relationships between them, and analyze spatiotemporal patterns over time. These detections can also serve as initial labeled data for ML research pertaining to the respective phenomena. The tool also incorporates feedback from subject matter experts to further improve the model detection accuracy, thereby facilitating human-in-the-loop. This presentation will provide an overview of the ML model development and cloud deployment. We also discuss the capabilities of the user interface for displaying the detections.

Muthukumaran Ramasubramanian↗

NASA NTRS1984

COSPAS-SARSAT Satellite Orbit Predictor. Volume 3

The satellite orbit predictor is a graphical aid for determining the relationship between the satellite (SARSAT or COSPAS) orbit, antenna coverage of the spacecraft and coverage of the LUTs. The predictor allows the user to quickly visualize if a selected position will probably be detected and is composed of a base map and a satellite track overlay for each satellite. Additionally, a table of equator crossings for each satellite is included.

Friedman, Morton L.↗

NASA NTRS1984

COSPAS-SARSAT Satellite Orbit Predictor, Vol 4

Friedman, Morton L.↗

NASA NTRS1987

COSPAS-SARSAT Satellite Orbit Predictor Volume 10

The satellite orbit predictor is a graphical aid for determining the relationship between the satellite (SARSAT or COSPAS) orbit, antenna coverage of the spacecraft and coverage of the Local User Terminals. The predictor allows the user to quickly visualize if a selected position will be detected and is composed of a base map and satellite track overlay for each satellite. A table of equator crossings for each satellite is included.

Friedman, Morton L.↗

NASA NTRS1975

Application of multispectral photography to mineral and land resources of South Carolina

The author has identified the following significant results. Good results were obtained from using Skylab photography in conjunction with LANDSAT imagery for visual interpretation of various geologic features, particularly lineaments. It was concluded that visual interpretation alone of Skylab photographs was quite limited, and much of this was because of the low contrast, heavily vegetated terrain in southeastern United States. Lineaments of major structural features are detectable but subtle. An intimate knowledge of the geologic field relationships is needed before a meaningful analysis is feasible using current satellite photography alone.

Olson, N. K.↗

DOE OSTI2023

Detecting lithium plating dynamics in a solid-state battery with operando X-ray computed tomography using machine learning

Operando X-ray micro-computed tomography (µCT) provides an opportunity to observe the evolution of Li structures inside pouch cells. Segmentation is an essential step to quantitatively analyzing µCT datasets but is challenging to achieve on operando Li-metal battery datasets due to the low X-ray attenuation of the Li metal and the sheer size of the datasets. Herein, we report a computational approach, batteryNET, to train an Iterative Residual U-Net-based network to detect Li structures. The resulting semantic segmentation shows singular Li-related component changes, addressing diverse morphologies in the dataset. In addition, visualizations of the dead Li are provided, including calculations about the volume and effective thickness of electrodes, deposited Li, and redeposited Li. We also report discoveries about the spatial relationships between these components. The approach focuses on a method for analyzing battery performance, which brings insight that significantly benefits future Li-metal battery design and a semantic segmentation transferrable to other datasets.

25 ENERGY STORAGE↗

NASA NTRS2019

Developing a Machine-Learning-Based Processing Framework for Twitter and Other Crowdsourced Data

Crowdsourced data streams such as Twitter and other social media are important sources of real-time and historical global information for Earth science applications. At the NASA Goddard Earth Sciences Data and Information Services Center (GES DISC), we have been exploring the Twitter data stream for its potential in augmenting the validation program of NASA's Global Precipitation Measurement (GPM) mission. To realize this potential, we need to increase the information density and enhance the quality of filtered precipitation tweets. We have implemented various components of a machine learning (ML)-based processing infrastructure for crowdsourced data that outputs, in this instance, useful and usable information derived from precipitation tweets. We have test enriched the Twitter stream with higher quality active tweets from those knowingly contributing to our effort and from existing crowdsourced programs (e.g., mPING, CoCoRaHS). We have experimented with various algorithms for processing tweets, including NaÃ ve Bayes, Convolutional Neural Network (CNN), Hierarchical Attention Network (HAN), and semi-supervised learning (with tri-training). Our current work focuses on (1) automated review of Earth science-related publications to determine relationships between discipline research needs and ML algorithms; (2) investigating Sequential Generative Adversarial Network (SeqGAN) for processing precipitation tweets for anomaly detection; and (3) managing crowdsourced data in a way that is compatible with existing NASA satellite data archives and using the data for ML applications. Key results include (1) network visualization of NLP-processed publications in various Earth science disciplines; (2) difference between GPM-linked, generated tweets and collected actual tweets that is small for GPM-determined light to moderate rain cases and high for GPM-determined heavy rain cases; and (3) identification of MongoDB for storing raw tweets and Zarr format for gridded tweets (compatible with GPM data). Our results have taken us a step closer to an operational ML-based tweet processing infrastructure and have already demonstrated that tweet-derived precipitation information is potentially useful for validation of Earth science satellite data.

Teng, William↗