Search NASA⌕ Search

SEARCH · Search NASA

Results for “Hyperspectral”

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 145 records · Page 8
All sourcesNASA NTRSDOE OSTI
NASA NTRS2011

The Hyperspectral Thermal Emission Spectrometer (HyTES): Preliminary Results

The Hyperspectral Thermal Emission Spectrometer (HyTES) is being developed as part of the risk reduction activities associated with the Hyperspectral Infrared Imager (HyspIRI). HyspIRI is one of the Tier 2 Decadal Survey Missions. HyTES will provide information on how to place the filters on the HyspIRI Thermal Infrared Instrument (TIR) as well as provide antecedent science data. The pushbroom design has 512 spatial pixels over a 50-degree field of view and 256 spectral channels between 7.5 micrometers to 12 micrometers. HyTES includes many key enabling state-of-the-art technologies including a high performance convex diffraction grating, a quantum well infrared photodetector (QWIP) focal plane array, and a compact Dyson-inspired optical design. The Dyson optical design allows for a very compact and optically fast system (F/1.6). It also minimizes cooling requirements due to the fact it has a single monolithic prism-like grating design which allows baffling for stray light suppression. The monolithic configuration eases mechanical tolerancing requirements which are a concern since the complete optical assembly is operated at cryogenic temperatures ((is) approximately 100K). The QWIP allows for optimum spatial and spectral uniformity and provides adequate responsivity or D-star to allow 200mK noise equivalent temperature difference (NEDT) operation across the LWIR passband. Assembly of the system is nearly complete. After completion, alignment results will be presented which show low keystone and smile distortion. This is required to minimize spatial-spectral mixing between adjacent spectral channels and spatial positions. Predictions show the system will have adequate signal to noise for laboratory calibration targets.

Moon Mineralogical Mapper (M3).↗
NASA NTRS2013

Space View Issues for Hyperspectral Sounders

The expectation for climate quality measurements from hyperspectral sounders is absolute calibration accuracy at the 100 mK level and stability at the < 40 mK/decade level. The Atmospheric InfraRed Sounder (AIRS)1, Cross-track Infrared Sounder (CrIS), and Infrared Atmospheric Sounding Interferometer (IASI) hyperspectral sounders currently in orbit have been shown to agree well over most of their brightness temperature range. Some larger discrepancies are seen, however, at the coldest scene temperatures, such as those seen in Antarctic winter and deep convective clouds. A key limiting factor for the calibrated scene radiance accuracy for cold scenes is how well the effective radiance of the cold space view pertains to the scene views. The space view signal is composed of external sources and instrument thermal emission at about 270 K from the scan mirror, external baffles, etc. Any difference in any of these contributions between space views and scene views will impact the absolute calibration accuracy, and the impact can be critical for cold scenes. Any change over time in these will show up as an apparent trend in calibrated radiances. We use AIRS data to investigate the validity of the space view assumption in view of the 100 mK accuracy and 40 mK/decade trend expectations. We show that the space views used for the cold calibration point for AIRS v5 Level-1B products meet these standards except under special circumstances and that AIRS v6 Level-1B products will meet them under all circumstances. This analysis also shows the value of having multiple distinct space views to give operational redundancy and analytic data, and that reaching climate quality requires continuing monitoring of aging instruments and adjustment of calibration.

trends↗
NASA NTRS2009

Fast and Adaptive Lossless On-Board Hyperspectral Data Compression System for Space Applications

Efficient on-board lossless hyperspectral data compression reduces the data volume necessary to meet NASA and DoD limited downlink capabilities. The techniques also improves signature extraction, object recognition and feature classification capabilities by providing exact reconstructed data on constrained downlink resources. At JPL a novel, adaptive and predictive technique for lossless compression of hyperspectral data was recently developed. This technique uses an adaptive filtering method and achieves a combination of low complexity and compression effectiveness that far exceeds state-of-the-art techniques currently in use. The JPL-developed 'Fast Lossless' algorithm requires no training data or other specific information about the nature of the spectral bands for a fixed instrument dynamic range. It is of low computational complexity and thus well-suited for implementation in hardware, which makes it practical for flight implementations of pushbroom instruments. A prototype of the compressor (and decompressor) of the algorithm is available in software, but this implementation may not meet speed and real-time requirements of some space applications. Hardware acceleration provides performance improvements of 10x-100x vs. the software implementation (about 1M samples/sec on a Pentium IV machine). This paper describes a hardware implementation of the JPL-developed 'Fast Lossless' compression algorithm on a Field Programmable Gate Array (FPGA). The FPGA implementation targets the current state of the art FPGAs (Xilinx Virtex IV and V families) and compresses one sample every clock cycle to provide a fast and practical real-time solution for Space applications.

image compression↗
NASA NTRS2009

Sparse Superpixel Unmixing for Exploratory Analysis of CRISM Hyperspectral Images

Fast automated analysis of hyperspectral imagery can inform observation planning and tactical decisions during planetary exploration. Products such as mineralogical maps can focus analysts' attention on areas of interest and assist data mining in large hyperspectral catalogs. In this work, sparse spectral unmixing drafts mineral abundance maps with Compact Reconnaissance Imaging Spectrometer (CRISM) images from the Mars Reconnaissance Orbiter. We demonstrate a novel "superpixel" segmentation strategy enabling efficient unmixing in an interactive session. Tests correlate automatic unmixing results based on redundant spectral libraries against hand-tuned summary products currently in use by CRISM researchers.

Image Segmentation↗
NASA NTRS2009

Hardware Implementation of Lossless Adaptive and Scalable Hyperspectral Data Compression for Space

On-board lossless hyperspectral data compression reduces data volume in order to meet NASA and DoD limited downlink capabilities. The technique also improves signature extraction, object recognition and feature classification capabilities by providing exact reconstructed data on constrained downlink resources. At JPL a novel, adaptive and predictive technique for lossless compression of hyperspectral data was recently developed. This technique uses an adaptive filtering method and achieves a combination of low complexity and compression effectiveness that far exceeds state-of-the-art techniques currently in use. The JPL-developed 'Fast Lossless' algorithm requires no training data or other specific information about the nature of the spectral bands for a fixed instrument dynamic range. It is of low computational complexity and thus well-suited for implementation in hardware. A modified form of the algorithm that is better suited for data from pushbroom instruments is generally appropriate for flight implementation. A scalable field programmable gate array (FPGA) hardware implementation was developed. The FPGA implementation achieves a throughput performance of 58 Msamples/sec, which can be increased to over 100 Msamples/sec in a parallel implementation that uses twice the hardware resources This paper describes the hardware implementation of the 'Modified Fast Lossless' compression algorithm on an FPGA. The FPGA implementation targets the current state-of-the-art FPGAs (Xilinx Virtex IV and V families) and compresses one sample every clock cycle to provide a fast and practical real-time solution for space applications.

FPGA implementation↗
NASA NTRS2014

Airborne Demonstration of FPGA Implementation of Fast Lossless Hyperspectral Data Compression System

Efficient on-board lossless hyperspectral data compression reduces data volume in order to meet NASA and DoD limited downlink capabilities. The technique also improves signature extraction, object recognition and feature classification capabilities by providing exact reconstructed data on constrained downlink resources. At JPL a novel, adaptive and predictive technique for lossless compression of hyperspectral data was recently developed. This technique uses an adaptive filtering method and achieves a combination of low complexity and compression effectiveness that far exceeds state-of-the-art techniques currently in use. The JPL-developed 'Fast Lossless' algorithm requires no training data or other specific information about the nature of the spectral bands for a fixed instrument dynamic range. It is of low computational complexity and thus well-suited for implementation in hardware.

data compression↗
NASA NTRS2014

Aerosol, Cloud and Trace Gas Observations Derived from Airborne Hyperspectral Radiance and Direct Beam Measurements in Recent Field Campaigns

The AERONET (AErosol RObotic NETwork) ground-based suite of sunphotometers provides measurements of spectral aerosol optical depth (AOD), precipitable water and spectral sky radiance, which can be inverted to retrieve aerosol microphysical properties that are critical to assessments of aerosol-climate interactions. Because of data quality criteria and sampling constraints, there are significant limitations to the temporal and spatial coverage of AERONET data and their representativeness for global aerosol conditions. The 4STAR (Spectrometer for Sky-Scanning, Sun-Tracking Atmospheric Research) instrument, jointly developed by NASA Ames and PNNL with NASA Goddard collaboration, combines airborne sun tracking and AERONET-like sky scanning with spectroscopic detection. Being an airborne instrument, 4STAR has the potential to fill gaps in the AERONET data set. Dunagan et al. [2013] present results establishing the performance of the instrument, along with calibration, engineering flight test, and preliminary scientific field data. The 4STAR instrument operated successfully in the SEAC4RS [Studies of Emissions and Atmospheric Composition, Clouds and Climate Coupling by Regional Surveys] experiment in Aug./Sep. 2013 aboard the NASA DC-8 and in the DoE [Department of Energy]-sponsored TCAP [Two Column Aerosol Project, July 2012 & Feb. 2013] experiment aboard the DoE G-1 aircraft (Shinozuka et al., 2013), and acquired a wealth of data in support of mission objectives on all SEAC4RS and TCAP research flights. 4STAR provided direct beam measurements of hyperspectral AOD, columnar trace gas retrievals (H2O, O3, NO2; Segal-Rosenheimer et al., 2014), and the first ever airborne hyperspectral sky radiance scans, which can be inverted to yield the same products as AERONET ground-based observations. In addition, 4STAR measured zenith radiances underneath cloud decks for retrievals of cloud optical depth and effective diameter. In this presentation, we provide an overview of the new 4STAR capabilities for airborne field campaigns, with an emphasis on comparisons between 4STAR and AERONET sky radiances, and retrievals of aerosol microphysical properties based on sky radiance measurements, column trace gas amounts from spectral direct beam measurements and cloud property retrievals from zenith mode observations for a few select case studies in the SEAC4RS and TCAP experiments. We summarize the aerosol, trace gas, cloud and airmass characterization studies made possible by the combined 4STAR direct beam, and sky/zenith radiance observations.

Aerosol-Climate↗
NASA NTRS2013

The Intercalibration of Geostationary Visible Imagers Using Operational Hyperspectral SCIAMACHY Radiances

Spectral band differences between sensors can complicate the process of intercalibration of a visible sensor against a reference sensor. This can be best addressed by using a hyperspectral reference sensor whenever possible because they can be used to accurately mitigate the band differences. This paper demonstrates the feasibility of using operational Scanning Imaging Absorption Spectrometer for Atmospheric Cartography (SCIAMACHY) large-footprint hyperspectral radiances to calibrate geostationary Earth-observing (GEO) sensors. Near simultaneous nadir overpass measurements were used to compare the temporal calibration of SCIAMACHY with Aqua Moderate Resolution Imaging Spectroradiometer band radiances, which were found to be consistent to within 0.44% over seven years. An operational SCIAMACHY/GEO ray-matching technique was presented, along with enhancements to improve radiance pair sampling. These enhancements did not bias the underlying intercalibration and provided enough sampling to allow up to monthly monitoring of the GEO sensor degradation. The results of the SCIAMACHY/GEO intercalibration were compared with other operational four-year Meteosat-9 0.65-μm calibration coefficients and were found to be within 1% of the gain, and more importantly, it had one of the lowest temporal standard errors of all the methods. This is more than likely that the GEO spectral response function could be directly applied to the SCIAMACHY radiances, whereas the other operational methods inferred a spectral correction factor. This method allows the validation of the spectral corrections required by other methods.

Doelling, David R.↗
NASA NTRS2015

An Active Learning Framework for Hyperspectral Image Classification Using Hierarchical Segmentation

Augmenting spectral data with spatial information for image classification has recently gained significant attention, as classification accuracy can often be improved by extracting spatial information from neighboring pixels. In this paper, we propose a new framework in which active learning (AL) and hierarchical segmentation (HSeg) are combined for spectral-spatial classification of hyperspectral images. The spatial information is extracted from a best segmentation obtained by pruning the HSeg tree using a new supervised strategy. The best segmentation is updated at each iteration of the AL process, thus taking advantage of informative labeled samples provided by the user. The proposed strategy incorporates spatial information in two ways: 1) concatenating the extracted spatial features and the original spectral features into a stacked vector and 2) extending the training set using a self-learning-based semi-supervised learning (SSL) approach. Finally, the two strategies are combined within an AL framework. The proposed framework is validated with two benchmark hyperspectral datasets. Higher classification accuracies are obtained by the proposed framework with respect to five other state-of-the-art spectral-spatial classification approaches. Moreover, the effectiveness of the proposed pruning strategy is also demonstrated relative to the approaches based on a fixed segmentation.

classification↗
NASA NTRS2019

Investigating the Utility of Hyperspectral Sounders in the 9.6 μm Band to Improve Ozone Analyses

Currently, hyperspectral sounder brightness temperatures assimilated in the Goddard Earth Observing System - Atmospheric Data Assimilation System (GEOS-ADAS) are limited to assimilating temperature and moisture. The ozone sensitive 9.6 micron region is sensed by several hyperspectral sounders including AIRS (Atmospheric InfraRed Sounder), IASI (Infrared Atmospheric Sounding Interferometer), and CrIS (Cross-track Infrared Sounder). Direct assimilation of brightness temperatures in the 9.6 micron region have been used previously to improve ozone analyses. This has recently been achieved by ECMWF (European Centre for Medium-Range Weather Forecast) (Dragani and McNally, 2013; Eresmaa et al., 2017), and while every system presents its challenges, it should be possible to take advantage of this spectral region using the GEOS-ADAS. For this study, channels were selected from available operational subsets evaluating information content, and minimizing inter-channel correlation. Additionally, information such as channel selections made by other studies, and vertical sensitivities of ozone and temperature were considered in developing the study. The analyses produced show improvements verified against ozonesondes taken from SHADOZ (Southern Hemisphere Additional Ozonesondes), and WOUDC (World Ozone and Ultraviolet Data Center). The addition of ozone channels does degrade forecast skill in the Tropics, on the border of statistical significance. Overall, the addition of these channels in some form could improve ozone analyses in the GEOS-ADAS.

Karpowicz, Bryan M.↗
NASA NTRS2019

Improvements to Ozone Analyses using Hyperspectral Sounders in the 9.6 μm Band

Previously, hyperspectral sounder brightness temperatures assimilated in the Goddard Earth Observing System Atmospheric Data Assimilation System (GEOS-ADAS) were limited to assimilating temperature and moisture. The ozone sensitive 9.6 m region is sensed by several hyperspectral sounders including AIRS (Atmospheric InfraRed Sounder), IASI (Infrared Atmospheric Sounding Interferometer), and CrIS (Cross-track Infrared Sounder). Direct assimilation of brightness temperatures in the 9.6 m region have been operational at ECMWF for several years (Dragani and McNally, 2013; Eresmaa et al., 2017). With this study, similar improvements using the GEOS-ADAS are presented. Channels were selected from available operational subsets evaluating information content and minimizing inter-channel correlation. Additionally, information such as channel selections made by other studies, and vertical sensitivities of ozone and temperature were considered in developing the study. The analyses produced show improvements verified against ozonesondes taken from SHADOZ (Southern Hemisphere Additional Ozonesondes). It is anticipated that inclusion of these ozone sensitive channels will be used to improve NASA GMAO products in the near future.

GEOS-ADAS↗
NASA NTRS2019

Improvements to Ozone Analyses Using Hyperspectral Sounders in the 9.6 μm Band

Previously, hyperspectral sounder brightness temperatures assimilated in the Goddard Earth Observing System Atmospheric Data Assimilation System (GEOS-ADAS) were limited to assimilating temperature and moisture. The ozone sensitive 9.6 m region is sensed by several hyperspectral sounders including AIRS (Atmospheric InfraRed Sounder), IASI (Infrared Atmospheric Sounding Interferometer), and CrIS (Cross-track Infrared Sounder). Direct assimilation of brightness temperatures in the 9.6 m region have been operational at ECMWF for several years (Dragani and McNally, 2013; Eresmaa et al., 2017). With this study, similar improvements using the GEOS-ADAS are presented. Channels were selected from available operational subsets evaluating information content and minimizing inter-channel correlation. Additionally, information such as channel selections made by other studies, and vertical sensitivities of ozone and temperature were considered in developing the study. The analyses produced show improvements verified against ozonesondes taken from SHADOZ (Southern Hemisphere Additional Ozonesondes), and WOUDC (World Ozone and Ultraviolet Data Center). Inclusion of inter-channel correlation error to the selected ozone channels are expected to improve ozone analyses further along with improve analyses overall. It is anticipated that inclusion of these ozone sensitive channels will be used to improve NASA GMAO products in the near future.

GEOS-ADAS↗
NASA NTRS2019

Improved Ozone Analyses Using Direct Assimilation of 9.6μm Radiances from Hyperspectral Sounders

Previously, hyperspectral sounder brightness temperatures assimilated in the Goddard Earth Observing System Atmospheric Data Assimilation System (GEOS-ADAS) were limited to assimilating temperature and moisture. The ozone sensitive 9.6 m region is sensed by several hyperspectral sounders including AIRS (Atmospheric InfraRed Sounder), IASI (Infrared Atmospheric Sounding Interferometer), and CrIS (Cross-track Infrared Sounder). Direct assimilation of brightness temperatures in the 9.6 m region have been operational at ECMWF for several years (Dragani and McNally, 2013; Eresmaa et al., 2017). With this study, similar improvements using the GEOS-ADAS are presented. Channels were selected from available operational subsets evaluating information content and minimizing inter-channel correlation. Additionally, information such as channel selections made by other studies, and vertical sensitivities of ozone and temperature were considered. The analyses produced show improvements verified against ozonesondes taken from SHADOZ (Southern Hemisphere Additional Ozonesondes), and WOUDC (World Ozone and Ultraviolet Data Center). While care was taken to minimize inter-channel correlation through channel selection, a key feature available in the GEOS-ADAS is the ability to account for correlated error. The importance of inter-channel correlated error is highlighted by performing assimilation experiments with and without inter-channel correlation in the GEOS-ADAS. It is anticipated that inclusion of these ozone sensitive channels will be used to improve NASA GMAO products in the near future.

Karpowicz, Bryan M.↗
NASA NTRS2020

Searching for Hyperspectral Optical Proxies to Aid Chesapeake Bay Resource Managers in the Detection of Poor Water Quality

Shellfish aquaculture is a growing industry in the Chesapeake Bay. As population grows near the coast, extreme weather events cause a greater volume of pollutant runoff from impervious surfaces and agricultural lands. Resource managers who monitor shellfish beds need reliable information on a variety of water quality indicators at higher frequency than is possible through field monitoring programs and at a higher level of detail than current satellite products can provide. Although many factors causing degraded water quality that can impact human health are not currently discernable by traditional multispectral techniques, hyperspectral imagery offers a new opportunity to detect phytoplankton communities associated with harmful algal blooms and biotoxin production. Together with resource managers in their routine monitoring of sites around the bay from small boats, we have been exploring remotely sensed optical proxies for the detection of harmful algal blooms and sewage. Early warning by remote sensing could guide sampling and improve the efficiency of shellfish bed closures, ultimately improving health outcomes for humans and animals. An extensive network of routine sampling by Chesapeake Bay Program managers makes this is an ideal location to develop and test future satellite data products to support management decisions. Next generation hyperspectral measurements from the future Plankton Aerosol Cloud ocean Ecosystem (PACE) mission at nearly daily frequency, combined with the potential of higher spatial resolution from the Surface Biology and Geology (SBG) observing system recommended in the recent Decadal Survey, along with high frequency observations from the newly selected Geostationary Littoral Imaging and Monitoring Radiometer (GLIMR) Earth Venture Instrument make this a critical time for defining the needs of the aquaculture and resource management community to save lives, time, and money.

Schollaert Uz, Stephanie↗
NASA NTRS2020

Synergistic Use of Hyperspectral UV-Visible OMI and Broadband Meteorological Imager MODIS Data for a Merged Aerosol Product

The retrieval of optimal aerosol datasets by the synergistic use of hyperspectral ultraviolet(UV)–visible and broadband meteorological imager (MI) techniques was investigated. The Aura Ozone Monitoring Instrument (OMI) Level 1B (L1B) was used as a proxy for hyperspectral UV–visible instrument data to which the Geostationary Environment Monitoring Spectrometer (GEMS) aerosol algorithm was applied. Moderate-Resolution Imaging Spectroradiometer (MODIS) L1B and dark target aerosol Level 2 (L2) data were used with a broadband MI to take advantage of the consistent time gap between the MODIS and the OMI. First, the use of cloud mask information from the MI infrared (IR) channel was tested for synergy. High-spatial-resolution and IR channels of the MI helped mask cirrus and sub-pixel cloud contamination of GEMS aerosol, as clearly seen in aerosol optical depth (AOD) validation with Aerosol Robotic Network (AERONET) data. Second, dust aerosols were distinguished in the GEMS aerosol-type classification algorithm by calculating the total dust confidence index (TDCI) from MODIS L1B IR channels. Statistical analysis indicates that the Probability of Correct Detection (POCD) between the forward and inversion aerosol dust models (DS) was increased from 72% to 94% by use of the TDCI for GEMS aerosol-type classification, and updated aerosol types were then applied to the GEMS algorithm. Use of the TDCI for DS type classification in the GEMS retrieval procedure gave improved single-scattering albedo (SSA) values for absorbing fine pollution particles (BC) and DS aerosols. Aerosol layer height (ALH) retrieved from GEMS was compared with Cloud-Aerosol Lidar with Orthogonal Polarization (CALIOP) data, which provides high-resolution vertical aerosol profile information. The CALIOP ALH was calculated from total attenuated backscatter data at 1064 nm, which is identical to the definition of GEMS ALH. Application of the TDCI value reduced the median bias of GEMS ALH data slightly. The GEMS ALH bias approximates zero, especially for GEMS AOD values of>~0.4 and GEMS SSA values of<~0.95.Finally, the AOD products from the GEMS algorithm and MI were used in aerosol merging with the maximum-likelihood estimation method, based on a weighting factor derived from the standard deviation of the original AOD products. With the advantage of the UV–visible channel in retrieving aerosol properties over bright surfaces, the combined AOD products demonstrated better spatial data availability than the original AOD products, with comparable accuracy. Furthermore, pixel-level error analysis of GEMS AOD data indicates improvement through MI synergy.

aerosol↗
NASA NTRS2018

Technology Development in Support of Hyperspectral Infrared Atmospheric Sounding in a CubeSat

Hyperspectral infrared sounding in a CubeSat will provide a new dimension to the current suite of IR sounders by allowing measurements at multiple times of day and enabling formation flying of IR sounders for new data products such as Atmospheric Motion Vector (AMV) winds. This paper focuses on technology development during the CubeSat Infrared Atmospheric Sounder (CIRAS) project sponsored by the NASA Earth Science Technology Office (ESTO), and coincident studies by the NOAA Office of Projects, Planning, and Analysis (OPPA). The CIRAS approach incorporates key new instrument technologies developed at JPL’s Microdevices Lab (MDL) including a 2D array of High Operating Temperature Barrier Infrared Detector (HOT-BIRD) material, selected for its high uniformity, low cost, low noise and higher operating temperatures than traditional materials. The second key technology is an MWIR Grating Spectrometer (MGS) designed by Ball Aerospace with a JPL MDL slit and immersion grating to provide hyperspectral infrared imaging in a CubeSat volume. The third key technology is a blackbody calibration target fabricated with MDL’s black silicon to have very high emissivity in a flat plate construction. JPL has completed design and breadboard of the mechanical, electronic and thermal subsystems for CIRAS payload including a HOT-BIRD FPA, with filters in a dewar and a breadboard of the electronics and scan mirror assembly. Blue Canyon Technologies, developer of the CIRAS 6U CubeSat, completed the Final Design Review for the spacecraft.

Yee, Karl↗
NASA NTRS2021

Investigating the Utility of Hyperspectral Sounders in the 9.6 μm Band to Improve Ozone Analyses

Hyperspectral sounder brightness temperatures assimilated in the Global Earth Observing System Atmospheric Data Assimilation System (GEOS-ADAS) were previously limited to assimilating temperature and moisture. The ozone-sensitive 9.6 𝜇 m region is sensed by several hyperspectral sounders including AIRS (Atmospheric InfraRed Sounder), IASI (Infrared Atmospheric Sounding Interferometer), and CrIS (Cross-track Infrared Sounder). Direct assimilation of brightness temperatures in the 9.6 𝜇 m region have been operational at ECMWF for several years. With this study, similar improvements using the GEOS-ADAS are presented. Channels were selected from available operational subsets evaluating information content and minimizing inter-channel correlation. Additionally, information such as channel selections made by other studies, and vertical sensitivities of ozone and temperature were considered. The analyses produced show improvements verified against ozonesondes taken from SHADOZ (Southern Hemisphere Additional Ozonesondes) and WOUDC (World Ozone and Ultraviolet Data Center). This work will be added to the GEOS-ADAS and will provide an improved source of ozone data in NASA Global Modeling and Assimilation Office products.

Ozone↗
NASA NTRSDate not supplied

Onboard Hyperspectral Image Classification via Transfer Learning for Communication-Limited Spacecraft

Employing deep-learning and artificial-intelligence (AI) techniques onboard spacecraft can dramatically improve priority data selection to ensure more effective use of the available downlink. However, deployment of effective deep-learning models requires significant training on the ground, which may not be feasible, due to limited data available in an unexplored environment. Therefore, this research explores building robust classification models for onboard data processing where training data is highly limited using transfer-learning techniques. In this paper, we focus on the use case of hyperspectral imaging for remote sensing, a domain where the high dimensionality of the data from the sensor can rapidly saturate the downlink bandwidth. With this bottleneck, there is an impending need to autonomously and robustly classify data onboard to optimize downlink of high-impact measurements, thus maximizing the scientific utility per bit transmitted to the ground. This paper examines the use of deep neural networks onboard for hyperspectral image classification in a communication-limited scenario to analyze how the models perform with limited training data. The use of transfer learning can ameliorate the issue of poor generalization by transferring features learned from training on a large source dataset for one classification task to the target classification task with limited training data. For two deep-learning models from literature, we compare the accuracy of the models trained using transfer learning to models trained from scratch using a random weight initialization with varying amounts of training data. We demonstrate the feasibility and performance of running inference of the deep-learning models on representative flight-like hardware.

Advanced Avionics, Machine Learning, Data Processi↗