Search NASA⌕ Search

SEARCH · Search NASA

Results for “change point 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.

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

At least 19 records

Amazonia Disasters: Assessing Methods for Gold Mining-related Deforestation Detection in Amazonia Using NASA Earth Observations

Artisanal and small-scale gold mining (ASGM) is responsible for a large fraction of deforestation and disturbance in Amazonia. These activities cause severe impacts on the rainforest ecosystem and socioeconomic state of the region. NASA DEVELOP partnered with the Asociación para la Conservación de la Cuenca Amazónica (ACCA), NASA SERVIR Science Coordination Office, and the Spatial Informatics Group to enhance ASGM-related deforestation detection methods. ACCA currently uses the Omnibus Q-test Change Point Detection Algorithm to identify changes in Synthetic Aperture Radar (SAR) monthly-aggregated temporal data from the Sentinel-1 satellite. The team determined the algorithm's accuracy by comparing a stratified random sample of change points against data from January 2019 to June 2020 identified using PlanetScope and Landsat 8 Operational Land Imager (OLI) Earth observations through Collect Earth Online. Our results indicated a users' accuracy of 55% for temporal change detection and producer's and user's accuracies of 99% and 97%, respectively, for detecting when change did not occur. Of the labeled change points, only 19% were due to mining activity. This research can help our partners have a more accurate understanding of where illegal gold mining may be taking place and inform decisions to remediate this activity.

DEVELOP Project Summary↗

Amazonia Disasters: Assessing Methods for Gold Mining-Related Deforestation Detection in Amazonia Using NASA Earth Observations

Artisanal and small-scale gold mining (ASGM) is responsible for a large fraction of deforestation and disturbance in Amazonia. These activities cause severe impacts on the rainforest ecosystem and socioeconomic state of the region. NASA DEVELOP partnered with the Asociación para la Conservación de la Cuenca Amazónica (ACCA), NASA SERVIR Science Coordination Office, and the Spatial Informatics Group to enhance ASGM-related deforestation detection methods. ACCA currently uses the Omnibus Q-test Change Point Detection Algorithm to identify changes in Synthetic Aperture Radar (SAR) monthly-aggregated temporal data from the Sentinel-1 satellite. The team determined the algorithm's accuracy by comparing a stratified random sample of change points against data from January 2019 to June 2020 identified using PlanetScope and Landsat 8 Operational Land Imager (OLI) Earth observations through Collect Earth Online. Our results indicated a users' accuracy of 55% for temporal change detection and producer's and user's accuracies of 99% and 97%, respectively, for detecting when change did not occur. Of the labeled change points, only 19% were due to mining activity. This research can help our partners have a more accurate understanding of where illegal gold mining may be taking place and inform decisions to remediate this activity.

DEVELOP Tech Paper↗

Variations in the Sea Ice Edge and the Marginal Ice Zone on Different Spatial Scales as Observed from Different Satellite Sensor

The Marginal sea Ice Zone (MIZ) and the sea ice edge are the most dynamic areas of the sea ice cover. Knowledge of the sea ice edge location is vital for routing shipping in the polar regions. The ice edge is the location of recurrent plankton blooms, and is the habitat for a number of animals, including several which are under severe ecological threat. Polar lows are known to preferentially form along the sea ice edge because of induced atmospheric baroclinicity, and the ice edge is also the location of both vertical and horizontal ocean currents driven by thermal and salinity gradients. Finally, sea ice is both a driver and indicator of climate change and monitoring the position of the ice edge accurately over long time periods enables assessment of the impact of global and regional warming near the poles. Several sensors are currently in orbit that can monitor the sea ice edge. These sensors, though, have different spatial resolutions, different limitations, and different repeat frequencies. Satellite passive microwave sensors can monitor the ice edge on a daily or even twice-daily basis, albeit with low spatial resolution - 25 km for the Special Sensor Microwave Imager (SSM/I) or 12.5 km for the Advanced Microwave Scanning Radiometer (AMSR-E). Although special methods exist that allow the detection of the sea ice edge at a quarter of that nominal resolution (PSSM). Visible and infrared data from the Advanced Very High Resolution Radiometer (AVHRR) and from the Moderate Resolution Imaging Spectroradiometer (MODIS) provide daily coverage at 1 km and 250 m, respectively, but the surface observations me limited to cloud-free periods. The Landsat 7 Enhanced Thematic Mapper (ETM+) has a resolution of 15 to 30 m but is limited to cloud-free periods as well, and does not provide daily coverage. Imagery from Synthetic Aperture Radar (SAR) instruments has resolutions of tens of meters to 100 m, and can be used to distinguish open water and sea ice on the basis of surface and volume scattering characteristics. The Canadian RADARSAT C-band SAR provides data that cover the Arctic Ocean and the MIZ every 3 days. A change-point detection approach was utilized to obtain an ice edge estimate from the RADARSAT data The Quickscat scatterometer provides ice edge information with a resolution of a few kilometers on a near-daily basis. During portions of March and April of 2003 a series of aircraft flights were conducted over the ice edge in the Bering Sea carrying the Polarimetric Scanning Radiometer (PSR), which provides spectral coverage identical with the AMSR-E instrument at a resolution of 500 meters. In this study we investigated these different data sets and analyzed differences in their definition of the sea ice edge and the marginal ice zone and how these differences as well as their individual limitations affect the monitoring of the ice edge dynamics. We also examined how the nature of the sea ice edge, including its location, compactness and shape, changes over the seasons. Our approach was based on calculation of distances between ice edges derived from the satellite and aircraft data sets listed above as well as spectral coherence methods and shape parameters such as tortuosity, curvature, and fractional dimension.

Markus, Thorsten↗

Unsupervised Change Detection for Space Habitats Using 3D Point Clouds

This work presents an algorithm for scene change detection from point clouds to enable autonomous robotic caretaking in future space habitats. Autonomous robotic systems will help maintain future deep-space habitats, such as the Gateway space station, which will be uncrewed for extended periods. Existing scene analysis software used on the International Space Station (ISS) relies on manually-labeled images for detecting changes. In contrast, the algorithm presented in this work uses raw, unlabeled point clouds as inputs. The algorithm first applies modified Expectation-Maximization Gaussian Mixture Model (GMM) clustering to two input point clouds. It then performs change detection by comparing the GMMs using the Earth Mover’s Distance. The algorithm is validated quantitatively and qualitatively using a test dataset collected by an Astrobee robot in the NASA Ames Granite Lab comprising single frame depth images taken directly by Astrobee and full-scene reconstructed maps built with RGB-D and pose data from Astrobee. The runtimes of the approach are also analyzed in depth. The source code is publicly released to promote further development.

robotics↗

ILRS Analysis Activities After the Adoption of ITRF2020

"The contribution of the International Laser Ranging Service (ILRS) to the most recent realization of the International Terrestrial Reference System (ITRS) was the result of an analysis strategy with two major modifications compared to the operational products: a modeling for stations long-term systematic errors (biases reported in the ILRS Data Handling File) and an updated model of the target signature error (satellite Centre of Mass model). Both refined models were used as input information for the ILRS contribution to the ITRF2020 (International Terrestrial Reference Frame 2020). Thereafter, the ILRS Analysis Standing Committee (ASC) focused its effort on implementing the new reference frame in its operational products, define a strategy to improve the ongoing monitoring of the systematic errors, compute the ILRS contribution to the planned ITRF2020 update, and to include LARES-2 among the considered satellites for the operational products. The ILRS ASC implemented the ITRF2020/SLRF2020 into all its official operational products (TRF, Earth Orientation Parameters, predicted and combined satellite orbits) and its impact was evaluated. The operational products benefit from the continuous monitoring of the station systematic errors and the frequent updates of the Data Handling File whenever a significant change in the station systematic error is observed. In the future, a change-point detection algorithm, jointly estimating the times and the number of discontinuities, will be implemented to detect potential new discontinuities in the range bias series. The inclusion of LARES-2 among the satellites whose data are operationally analyzed will furtherly increase the robustness of the estimated parameters. Finally, the ILRS ASC activities include the benchmarking of a new analysis center (CNES) which will formally begin its own contribution in 2024."

Satellite Laser Ranging↗

Method for Measuring Collimator-Pointing Sensitivity to Temperature Changes

For a variety of applications, it is important to measure the sensitivity of the pointing of a beam emerging from a collimator, as a function of temperature changes. A straightforward method for carrying out this measurement is based on using interferometry for monitoring the changes in beam pointing, which presents its own problems. The added temperature dependence and complexity issues relating to using an interferometer are addressed by not using an interferometer in the first place. Instead, the collimator is made part of an arrangement that uses a minimum number of low-cost, off-the-shelf materials and by using a quad diode to measure changes in beam pointing. In order to minimize the influence of the test arrangement on the outcome of the measurement, several steps are taken. The collimator assembly is placed on top of a vertical, 1-m-long, fused silica tube. The quad diode is bonded to a fused silica bar, which, in turn, is bonded to the lower end of the fused silica tube. The lower end of the tube rests on a self-aligning support piece, while the upper end of the tube is kept against two rounded setscrew tips, using a soft rubber string. This ensures that very little stress is applied to the tube as the support structure changes dimensions due to thermal expansion. Light is delivered to the collimator through a bare fiber in order to minimize variable bending torque caused by a randomly relaxing, rigid fiber jacket. In order to separate the effect of temperature on the collimator assembly from the effect temperature has on the rest of the setup, multiple measurements are taken with the collimator assembly rotated from measurement to measurement. Laboratory testing, with 1-m spacing between the collimator and the quad diode, has shown that the sensitivity of the arrangement is better than 100 nm rms, over time spans of at least one hour, if the beam path is protected from atmospheric turbulence by a tube. The equivalent sensitivity to detecting changes in pointing angle is 100 nanoradians.

Abramovici, Alex↗

Ocular Outcomes Comparison Between 14- and 70-day Head-down Tilt Bed Rest

INTRODUCTION: Ophthalmological changes, including optic disc edema with optic nerve sheath distension, posterior globe flattening with hyperopic shift, choroidal folds and cotton wool spots have been detected in some astronauts involved in long-duration spaceflights. (sup 1) It is hypothesized that elevated intracranial pressure resulting from microgravity-induced cephalad fluid shifts may be responsible for most of these findings. Head-down tilt bed rest (HTDBR) is a ground-based microgravity analog which also produces cephalad fluid shifts. It is conceivable that prolonged HDTBR exposure may induce ocular changes similar to those experienced in microgravity. Therefore, the purpose of the present study was to compare structural and functional ocular outcomes between 14- and 70-day HDTBR in healthy human subjects. It is hypothesized that 70-d HDTBR induced ocular changes of greater magnitude as compared to 14-d HDTBR. METHODS: Two HDTBR studies were conducted at the NASA Flight Analogs Research Unit, located at The University of Texas Medical Branch at Galveston, TX. Identical NASA standard screening procedures and BR conditions (e.g., strict sleep-wake cycle, standardized diet, continuous video monitoring) were implemented in both studies. Participants spent 14 and/or 70 consecutive days in a 6deg HDT position and did not engage in exercise. Subjects received weekly ocular examinations before, during, and after HDTBR. Ocular testing included: distance and near best-corrected visual acuity (BCVA), cycloplegic refraction, intraocular pressure (IOP) measurement, color vision, red dot test, modified Amsler grid test, confrontational visual field, color fundus photography and Spectral-domain OCT scans of the macula and the optic disc. Pre/post HDTBR differences between the two studies will be evaluated for BCVA, spherical equivalent, IOP, retinal nerve fiber layer (RNFL) thickness and macular OCT parameters. RESULTS: 16 (12 males and 4 females) and 6 (5 males and 1 female) subjects participated in the 14 and 70-day HDTBR studies, respectively. One subject participated in both studies. The demographic and ophthalmological characteristics of the 14-day HDTBR study have been presented elsewhere.2 In the 70- day study, subjects age averaged 39.5 +/- 7.8 years; pre- and post-HDTBR Goldmann IOP were, on average, 15.4 and 14.6 mmHg in the right eye, and 15.3 and 14.4 mmHg in the left eye, respectively; preand post-HDTBR Spectralis OCT (Heidelberg Engineering, GmbH, Heidelberg, Germany) average RNFL thickness were 103.8 and 106 μm in the right eye, and 103 and 104.3 μm in the left eye, respectively. In both studies, color vision, red dot test, modified Amsler grid test and confrontational visual field were within normal limits at all time points; no changes were detected on stereoscopic color optic disc photography. For the above outcomes, preliminary results of the pre/post HDTBR differences between the two studies will be presented at the 2015 Human Research Program Investigators' Workshop. CONCLUSIONS: Although 14-day HDTBR did not seem to induce clinically relevant ocular changes, (sup 2) a systematic evaluation and comparison with the 70-day HDTBR study results will elucidate whether the magnitude of ocular structural and functional changes is affected by the time spent in the recumbent position. Further research will also address the effects of integrated resistance and aerobic training during HDTBR on the ocular outcomes. REFERENCES: 1. Mader TH et al. Optic disc edema, globe flattening, choroidal folds, and hyperopic shifts observed in astronauts after long-duration space flight. Ophthalmology. 2011;118(10):2058-69. 2. Taibbi G et al. Ocular outcomes evaluation in a 14-day head-down bed rest study. Aviat Space Environ Med. 2014;85(10):983-92.

Cromwell, R.L.↗

Utilization of ERTS-1 data in the Houston area

The author has identified the following significant results. Using clustering techniques, several large lakes in Texas have been accurately delineated in computer printout graymaps. It was also found that small bodies of water (one to two acres in size) could be detected by searching for small reflectance values in the infrared data. A graymap printout of a lake described a shore outline that was not consistent with available maps. Field examination revealed that the actual level of the lake was below that for which the map was drawn. The current lake configuration agrees in shape and relative size with the ERTS-1 data printout. Water turbidity causes reflectance changes which are detectable in ERTS-1 band 7 data. A comparison has been made of the Monterey Bay, California area using 1971 aerial color infrared photography and a 1972 ERTS-1 band 7 infrared image. This comparison revealed that some event has occurred to impound a significant amount of water in the area since the infrared photography was taken. Data values in the ERTS-1 infrared image exhibit detectable changes in brightness at inflow points, where high turbidity would be present. Researchers had not expected to detect water turbidity patterns in band 7 (800 to 1100 nanometers).

Erb, R. B.↗

Analysis of ICESat Data Using Kalman Filter and Kriging to Study Height Changes in East Antarctica

We analyze ICESat derived heights collected between Feb. 03-Nov. 04 using a kriging/Kalman filtering approach to investigate height changes in East Antarctica. The model's parameters are height change to an a priori static digital height model, seasonal signal expressed as an amplitude Beta and phase Theta, and height-change rate dh/dt for each (100 km)(exp 2) block. From the Kalman filter results, dh/dt has a mean of -0.06 m/yr in the flat interior of East Antarctica. Spatially correlated pointing errors in the current data releases give uncertainties in the range 0.06 m/yr, making height change detection unreliable at this time. Our test shows that when using all available data with pointing knowledge equivalent to that of Laser 2a, height change detection with an accuracy level 0.02 m/yr can be achieved over flat terrains in East Antarctica.

Herring, Thomas A.↗

Ten Years of Land Cover Change on the California Coast Detected using Landsat Satellite Image Analysis

Landsat satellite imagery was analyzed to generate a detailed record of 10 years of vegetation disturbance and regrowth for Pacific coastal areas of Marin and San Francisco Counties. The Landsat Ecosystem Disturbance Adaptive Processing System (LEDAPS) methodology, a transformation of Tasseled-Cap data space, was applied to detected changes in perennial coastal shrubland, woodland, and forest cover from 1999 to 2009. Results showed several principal points of interest, within which extensive contiguous areas of similar LEDAPS vegetation change (either disturbed or restored) were detected. Regrowth areas were delineated as burned forest areas in the Point Reyes National Seashore (PRNS) from the 1995 Vision Fire. LEDAPS-detected disturbance patterns on Inverness Ridge, PRNS in areas observed with dieback of tanoak and bay laurel trees was consistent with defoliation by sudden oak death (Phytophthora ramorum). LEDAPS regrowth pixels were detected over much of the predominantly grassland/herbaceous cover of the Olema Valley ranchland near PRNS. Extensive restoration of perennial vegetation cover on Crissy Field, Baker Beach and Lobos Creek dunes in San Francisco was identified. Based on these examples, the LEDAPS methodology will be capable of fulfilling much of the need for continual, low-cost monitoring of emerging changes to coastal ecosystems.

Land Cover Change↗

Ultrasonic attenuation as an indicator of fatigue life of graphite/epoxy fiber composite

The narrow band ultrasonic longitudinal wave velocity and attenuation were measured as a function of the transfiber compression-compression fatigue of unidirectional graphite/epoxy composites. No change in velocity was detected at any point in fatigue life. For specimens fatigued at 80% of static strength, there was generally a 5% to 10% increase in attenuation, however, this increase does not appear to be a satisfactory indicator of fatigue life. On the other hand, there appears to be a correlation between initial attenuation (measured prior to cycling) and cycles to fracture. Initial attenuation as measured at 1.5 MHz and 2.0 MHz appears to be a good indicator of relative fatigue life.

Williams, J. H., Jr.↗

Identification of Health Events in Astronaut Missions Using Longitudinal Molecular Signature Detection

Individualized health monitoring can now incorporate a precision medicine approach, profiling multiple molecular and physiological measures of health (generalized omics) longitudinally to enable the timely diagnosis and treatment of disease. Such measurements can include blood chemistries, gene expression data, metabolite measurements, and digital device data. We will present our work on extending such an approach to monitoring individual astronaut health for deep space missions. We have developed and implemented novel algorithms to monitor and detect physiolgical state departures from individualized healthy astronaut baselines , utilizing and biologically annotating generalized omics. Our new methods can detect baseline deviations across omics corresponding to potentially adverse medical events. Events pointing to changes in individual health are then compared across individuals to identify common responses and detect changes affecting multiple crewmembers. We show the utility of our methods in detecting temporal health changes across subjects using retrospective Earth and astronaut mission data (metabolite and immune marker data across multiple missions), in order for this technique t o be applicable for future missions.

G I Mias↗

NASA systems autonomy demonstration project: Advanced automation demonstration of Space Station Freedom thermal control system

The NASA Systems Autonomy Demonstration Project (SADP) was initiated in response to Congressional interest in Space station automation technology demonstration. The SADP is a joint cooperative effort between Ames Research Center (ARC) and Johnson Space Center (JSC) to demonstrate advanced automation technology feasibility using the Space Station Freedom Thermal Control System (TCS) test bed. A model-based expert system and its operator interface were developed by knowledge engineers, AI researchers, and human factors researchers at ARC working with the domain experts and system integration engineers at JSC. Its target application is a prototype heat acquisition and transport subsystem of a space station TCS. The demonstration is scheduled to be conducted at JSC in August, 1989. The demonstration will consist of a detailed test of the ability of the Thermal Expert System to conduct real time normal operations (start-up, set point changes, shut-down) and to conduct fault detection, isolation, and recovery (FDIR) on the test article. The FDIR will be conducted by injecting ten component level failures that will manifest themselves as seven different system level faults. Here, the SADP goals, are described as well as the Thermal Control Expert System that has been developed for demonstration.

Dominick, Jeffrey↗

Application of machine learning and expert systems to Statistical Process Control (SPC) chart interpretation

Statistical Process Control (SPC) charts are one of several tools used in quality control. Other tools include flow charts, histograms, cause and effect diagrams, check sheets, Pareto diagrams, graphs, and scatter diagrams. A control chart is simply a graph which indicates process variation over time. The purpose of drawing a control chart is to detect any changes in the process signalled by abnormal points or patterns on the graph. The Artificial Intelligence Support Center (AISC) of the Acquisition Logistics Division has developed a hybrid machine learning expert system prototype which automates the process of constructing and interpreting control charts.

Shewhart, Mark↗

Sensitivity of N2O, CFCl3, and CF2Cl2 two-dimensional distributions to O2 absorption cross sections

A good test of the transport and ultraviolet radiation treatment for any model can be based on the consideration of the three gases N2O, CFCl3 (F-11), and CF2Cl2 (F-12). Attention is given to a two-dimensional model which makes it possible to study changes in seasonal and latitudinal distributions which result from O2 cross-section changes. A latitude-dependent ozone decrease has already been predicted by two-dimensional models for chlorofluoromethane (CFM) release. For this reason, it is of interest to determine if a strong latitude-dependent change is detectable in the CFM profiles. One-dimensional models have pointed toward a trend of decreased trace gas concentrations in correlation with decreased O2 cross sections. The O2 cross sections between 180 and 230 nm are changed within a range of experimentlly allowed values. It is found that lower O2 absorption cross sections lead to larger photodissociation rates for the trace gases F-11, F-12, and N2O.

Jackman, C. H.↗

Experimental results for the rapid determination of the freezing point of fuels

Two methods for the rapid determination of the freezing point of fuels were investigated: an optical method, which detected the change in light transmission from the disappearance of solid particles in the melted fuel; and a differential thermal analysis (DTA) method, which sensed the latent heat of fusion. A laboratory apparatus was fabricated to test the two methods. Cooling was done by thermoelectric modules using an ice-water bath as a heat sink. The DTA method was later modified to eliminate the reference fuel. The data from the sample were digitized and a point of inflection, which corresponds to the ASTM D-2386 freezing point (final melting point), was identified from the derivative. The apparatus was modifified to cool the fuel to -60 C and controls were added for maintaining constant cooling rate, rewarming rate, and hold time at minimum temperature. A parametric series of tests were run for twelve fuels with freezing points from -10 C to -50 C, varying cooling rate, rewarming rate, and hold time. Based on the results, an optimum test procedure was established. The results showed good agreement with ASTM D-2386 freezing point and differential scanning calorimetry results.

Mathiprakasam, B.↗