Hyperspectral sensor systems data sources for enhanced analysis
Explore the source record for details and available documents.
SEARCH · Engineering Papers
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.
Explore the source record for details and available documents.
The Sensor Intercomparison and Merger for Biological and Interdisciplinary Oceanic Studies (SIMBIOS) Project is concerned with ocean color satellite sensor data intercomparison and merger for biological and interdisciplinary studies of the global oceans. Imagery from different ocean color sensors can now be processed by a single software package using the same algorithms, adjusted by different sensor spectral characteristics, and the same ancillary meteorological and environmental data. This enables cross-comparison and validation of the data derived from satellite sensors and, consequently, creates continuity in ocean color information on both the temporal and spatial scale. The next step in this process is the integration of in situ ocean and atmospheric parameters to enable cross-validation and further refinement of the ocean color methodology. The SIMBIOS Project Office accomplishments during 2000 year are summarized under satellite data processing, data product validation, SeaWiFS Bio-Optical Archive and Storage System (SeaBASS) database, supporting services, sun photometers and calibration activities, and calibration round robins. These accomplishments are described.
Over the past 12 years, large volumes of precipitation data have been generated from space-based observatories (e.g., TRMM), merging of data products (e.g., gridded 3B42), models (e.g., GMAO), climatologies (e.g., Chang SSM/I derived rain indices), field campaigns, and ground-based measuring stations. The science research, applications, and education communities have greatly benefited from the unrestricted availability of these data from the Goddard Earth Sciences Data and Information Services Center (GES DISC) and, in particular, the services tailored toward precipitation data access and usability. In addition, tools and services that are responsive to the expressed evolving needs of the precipitation data user communities have been developed at the Precipitation Data and Information Services Center (PDISC) (http://disc.gsfc.nasa.gov/precipitation or google NASA PDISC), located at the GES DISC, to provide users with quick data exploration and access capabilities. In recent years, data management and access services have become increasingly sophisticated, such that they now afford researchers, particularly those interested in multi-data set science analysis and/or data validation, the ability to homogenize data sets, in order to apply multi-variant, comparison, and evaluation functions. Included in these services is the ability to capture data quality and data provenance. These interoperability services can be directly applied to future data sets, such as those from the Global Precipitation Measurement (GPM) mission. This presentation describes the data sets and services at the PDISC that are currently used by precipitation science and applications researchers, and which will be enhanced in preparation for GPM and associated multi-sensor data research. Specifically, the GES-DISC Interactive Online Visualization ANd aNalysis Infrastructure (Giovanni) will be illustrated. Giovanni enables scientific exploration of Earth science data without researchers having to perform the complicated data access and match-up processes. In addition, PDISC tool and service capabilities being adapted for GPM data will be described, including the Google-like Mirador data search and access engine; semantic technology to help manage large amounts of multi-sensor data and their relationships; data access through various Web services (e.g., OPeNDAP, GDS, WMS, WCS); conversion to various formats (e.g., netCDF, HDF, KML (for Google Earth)); visualization and analysis of Level 2 data profiles and maps; parameter and spatial subsetting; time and temporal aggregation; regridding; data version control and provenance; continuous archive verification; and expertise in data-related standards and interoperability. The goal of providing these services is to further the progress towards a common framework by which data analysis/validation can be more easily accomplished.
Remote sensing methods used to generate base maps to analyze the urban environment rely predominantly on digital sensor data from space-borne platforms. This is due in part from new sources of high spatial resolution data covering the globe, a variety of multispectral and multitemporal sources, sophisticated statistical and geospatial methods, and compatibility with GIS data sources and methods. The goal of this chapter is to review the four groups of classification methods for digital sensor data from space-borne platforms; per-pixel, sub-pixel, object-based (spatial-based), and geospatial methods. Per-pixel methods are widely used methods that classify pixels into distinct categories based solely on the spectral and ancillary information within that pixel. They are used for simple calculations of environmental indices (e.g., NDVI) to sophisticated expert systems to assign urban land covers. Researchers recognize however, that even with the smallest pixel size the spectral information within a pixel is really a combination of multiple urban surfaces. Sub-pixel classification methods therefore aim to statistically quantify the mixture of surfaces to improve overall classification accuracy. While within pixel variations exist, there is also significant evidence that groups of nearby pixels have similar spectral information and therefore belong to the same classification category. Object-oriented methods have emerged that group pixels prior to classification based on spectral similarity and spatial proximity. Classification accuracy using object-based methods show significant success and promise for numerous urban 3 applications. Like the object-oriented methods that recognize the importance of spatial proximity, geospatial methods for urban mapping also utilize neighboring pixels in the classification process. The primary difference though is that geostatistical methods (e.g., spatial autocorrelation methods) are utilized during both the pre- and post-classification steps. Within this chapter, each of the four approaches is described in terms of scale and accuracy classifying urban land use and urban land cover; and for its range of urban applications. We demonstrate the overview of four main classification groups in Figure 1 while Table 1 details the approaches with respect to classification requirements and procedures (e.g., reflectance conversion, steps before training sample selection, training samples, spatial approaches commonly used, classifiers, primary inputs for classification, output structures, number of output layers, and accuracy assessment). The chapter concludes with a brief summary of the methods reviewed and the challenges that remain in developing new classification methods for improving the efficiency and accuracy of mapping urban areas.
The subsonic stability and control testing of the Space Shuttle Orbiter in its two test flights in the tailcone-off configuration is discussed, and test results are presented. Flight test maneuvers were designed to maximize the quality and quantity of stability and control data in the minimal time allotted using the Space Shuttle Functional Simulator and the Modified Maximum Likelihood Estimator (MMLE) programs, and coefficients were determined from standard sensor data sets using the MMLE, despite problems encountered in timing due to the different measurement systems used. Results are included for lateral directional and longitudinal maneuvers as well as the Space Shuttle aerodynamic data base obtained using the results of wind tunnel tests. The flight test data are found to permit greater confidence in the data base since the differences found are well within control system capability. It is suggested that the areas of major differences, including lateral directional data with open speedbrake, roll due to rudder and normal force due to elevon, be investigated in any further subsonic flight testing. Improvements in sensor data and data handling techniques for future orbital test flights are indicated.
Sensor nodes composed of three basic components... radio module: COTS radio module implementing standardized WSN protocol; treated as WSN modem by main board main board: contains application processor (TI MSP430 microcontroller), memory, power supply; responsible for sensor data acquisition, pre-processing, and task scheduling; re-used in every application with growing library of embedded C code sensor card: contains application-specific sensors, data conditioning hardware, and any advanced hardware not built into main board (DSPs, faster A/D, etc.); requires (re-) development for each application.
The problem of how the scale, or spatial resolution, of reflectance data impacts retrievals of vegetation leaf area index (LAI) and fraction absorbed photosynthetically active radiation (PAR) has been investigated. We define the goal of scaling as the process by which it is established that LAI and FPAR values derived from coarse resolution sensor data equal the arithmetic average of values derived independently from fine resolution sensor data. The increasing probability of land cover mixtures with decreasing resolution is defined as heterogeneity, which is a key concept in scaling studies. The effect of pixel heterogeneity on spectral reflectances and LAI/FPAR retrievals is investigated with 1 km Advanced Very High Resolution Radiometer (AVHRR) data aggregated to different coarse spatial resolutions. It is shown that LAI retrieval errors at coarse resolution are inversely related to the proportion of the dominant land cover in such pixel. Further, large errors in LAI retrievals are incurred when forests are minority biomes in non-forest pixels compared to when forest biomes are mixed with one another, and vice-versa. A physically based technique for scaling with explicit spatial resolution dependent radiative transfer formulation is developed. The successful application of this theory to scaling LAI retrievals from AVHRR data of different resolutions is demonstrated
Detecting clouds within a satellite image is essential for retrieving surface geophysical parameters, such as albedo and temperature, from optical and thermal imagery because the retrieval methods tend to be valid for clear skies only. Thus, routine satellite data processing requires reliable automated cloud detection algorithms that are applicable to many surface types. Unfortunately, cloud detection over snow and ice is difficult due to the lack of spectral contrast between clouds and snow. Snow and clouds are both highly reflective in the visible wavelen,ats and often show little contrast in the thermal Infrared. However, at 1.6 microns, the spectral signatures of snow and clouds differ enough to allow improved snow/ice/cloud discrimination. The recent Terra and Aqua Moderate Resolution Imaging Spectro-Radiometer (MODIS) sensors have a channel (channel 6) at 1.6 microns. Presently the most comprehensive, long-term information on surface albedo and temperature over snow- and ice-covered surfaces comes from the Advanced Very High Resolution Radiometer ( AVHRR) sensor that has been providing imagery since July 1981. The earlier AVHRR sensors (e.g. AVHRR/2) did not however have a channel designed for discriminating clouds from snow, such as the 1.6 micron channel available on the more recent AVHRR/3 or the MODIS sensors. In the absence of the 1.6 micron channel, the AVHRR Polar Pathfinder (APP) product performs cloud detection using a combination of time-series analysis and multispectral threshold tests based on the satellite's measuring channels to produce a cloud mask. The method has been found to work reasonably well over sea ice, but not so well over the ice sheets. Thus, improving the cloud mask in the APP dataset would be extremely helpful toward increasing the accuracy of the albedo and temperature retrievals, as well as extending the time-series of albedo and temperature retrievals from the more recent sensors to the historical ones. In this work, we use data mining methods to construct a model of MODIS channel 6 as a function of other channels that are common to both MODIS and AVHRR. The idea is to use the model to generate the equivalent of MODIS channel 6 for AVHRR as a function of the AVHRR equivalents to MODIS channels. We call this a Virtual Sensor because it predicts unmeasured spectra. The goal is to use this virtual channel 6. to yield a cloud mask superior to what is currently used in APP . Our results show that several data mining methods such as multilayer perceptrons (MLPs), ensemble methods (e.g., bagging), and kernel methods (e.g., support vector machines) generate channel 6 for unseen MODIS images with high accuracy. Because the true channel 6 is not available for AVHRR images, we qualitatively assess the virtual channel 6 for several AVHRR images.
With the tragic passing this year of Gregory Leptoukh, the Earth and Space Sciences community lost a tireless participant in--and advocate for--science informatics. Throughout his career at NASA, Dr. Leptoukh established a theme of bridging the gulf between the informatics and science communities. Nowhere is this more evident than his leadership in the development of Giovanni (GES DISC Interactive Online Visualization ANd aNalysis Infrastructure). Giovanni is an online tool that serves to hide the often-complex technical details of data format and structure, making science data easier to explore and use by Earth scientists. To date Giovanni has been acknowledged as a contributor in 500-odd scientific articles. In recent years, Leptoukh concentrated his efforts on multi-sensor data inter-comparison, merging and fusion. This work exposed several challenges at the intersection of data and science. One of these was the ease with which a naive user might generate spurious comparisons, a potential hazard that was the genesis of the Multi-sensor Data Synergy Advisor (MDSA). The MDSA uses semantic ontologies and inference rules to organize knowledge about dataset quality and other salient characteristics in order to advise users on potential caveats for comparing or merging two datasets. Recently, Leptoukh also led the development of AeroStat, an online Giovanni instance to investigate aerosols via statistics from station and satellite comparisons and merged maps of data from more than one instrument. Aerostat offers a neural net based bias adjustment to harmonize the data by removing systematic offsets between datasets before merging. These examples exhibit Leptoukh's talent for adopting advanced computer technologies in the service of making science data more accessible to researchers. In this, he set an example that is at once both vital and challenging for the ESSI community to emulate.
The National Aeronautics and Space Administration (NASA) Space Station Program presents new opportunities for the application of telerobotic and robotic systems. The Laboratory Telerobotic Manipulator (LTM) is a highly advanced 7 degrees-of-freedom (DOF) telerobotic/robotic manipulator. It was developed and built for the Automation Technology Branch at NASA's Langley Research Center (LaRC) for work in research and to demonstrate ground-based telerobotic manipulator system hardware and software systems for future NASA applications in the hazardous environment of space. The LTM manipulator uses an embedded wiring design with all electronics, motor power, and control and communication cables passing through the pitch-yaw differential joints. This design requires the number of cables passing through the pitch/yaw joint to be kept to a minimum. To eliminate the cables needed to carry each pitch-yaw joint's sensor data to the VME control computers, a custom-embedded electronics package for each manipulator joint was developed. The electronics package collects and sends the joint's sensor data to the VME control computers over a fiber optic cable. The electronics package consist of five individual subsystems: the VME Link Processor, the Joint Processor and the Joint Processor power supply in the joint module, the fiber optics communications system, and the electronics and motor power cabling.
An expert system is being developed which can detect anomalies in Space Shuttle Main Engine (SSME) sensor data significantly earlier than the redline algorithm currently in use. The training of such an expert system focuses on two approaches which are based on low frequency and high frequency analyses of sensor data. Both approaches are being tested on data from SSME tests and their results compared with the findings of NASA and Rocketdyne experts. Prototype implementations have detected the presence of anomalies earlier than the redline algorithms that are in use currently. It therefore appears that these approaches have the potential of detecting anomalies early eneough to shut down the engine or take other corrective action before severe damage to the engine occurs.
This paper describes a quantitative methodology for bounding the false positive (FP) and false negative (FN) probabilities associated with a human-rated launch vehicle abort trigger (AT) that includes sensor data qualification (SDQ). In this context, an AT is a hardware and software mechanism designed to detect the existence of a specific abort condition. Also, SDQ is an algorithmic approach used to identify sensor data suspected of being corrupt so that suspect data does not adversely affect an AT's detection capability. The FP and FN methodologies presented here were developed to support estimation of the probabilities of loss of crew and loss of mission for the Space Launch System (SLS) which is being developed by the National Aeronautics and Space Administration (NASA). The paper provides a brief overview of system health management as being an extension of control theory; and describes how ATs and the calculation of FP and FN probabilities relate to this theory. The discussion leads to a detailed presentation of the FP and FN methodology and an example showing how the FP and FN calculations are performed. This detailed presentation includes a methodology for calculating the change in FP and FN probabilities that result from including SDQ in the AT architecture. To avoid proprietary and sensitive data issues, the example incorporates a mixture of open literature and fictitious reliability data. Results presented in the paper demonstrate the effectiveness of the approach in providing quantitative estimates that bound the probability of a FP or FN abort determination.
There is evidence in breast cancer and other malignancies that the physiologic environment within a tumor correlates with clinical outcome. We are developing a unique percutaneous Smart Probe to be used at the time of needle biopsy of the breast. The Smart Probe will simultaneously measure multiple physiologic parameters within a breast tumor. Direct and indirect measurements of tissue oxygen levels, blood flow, pH, and tissue fluid pressure will be analyzed in real-time. These parameters will be interpreted individually and collectively by innovative neural network techniques using advanced intelligent software. The goals are 1) develop a pecutaneous Smart Probe with multiple sensor modalities and applying advanced Information Technologies to provide real time diagnostic information of the tissue at tip of the probe, 2) test the percutaneous Smart Probe in women with benign and malignant breast masses who will be undergoing surgical biopsy, 3) correlate probe sensor data with benign and malignant status of breast masses, 4) determine whether the probe can detect physiologic differences within a breast tumor, and its margins, and in adjacent normal breast tissue, 5) correlate probe sensor data with known prognostic factors for breast caner, including tumor size, tumor grade, axillary lymph node metastases, estrogen receptor and progesterone receptor status.
In this paper, we describe a method of providing guaranteed delivery of sensor data gathered at arbitrary times using an only intermittently available radio frequency identification (RFID) transport scheme. This technique provides a passive (e.g., non-powered) interface for a power-constrained embedded device to transport sensor data while overcoming the limitation of infrequent access to an RFID interrogator.
This invention relates to a reconfigurable fuzzy cell comprising a digital control programmable gain operation amplifier, an analog-to-digital converter, an electrically erasable PROM, and 8-bit counter and comparator, and supporting logic configured to achieve in real-time fuzzy systems high throughput, grade-of-membership or membership-value conversion of multi-input sensor data. The invention provides a flexible multiplexing-capable configuration, implemented entirely in hardware, for effectuating S-, Z-, and PI-membership functions or combinations thereof, based upon fuzzy logic level-set theory. A membership value table storing 'knowledge data' for each of S-, Z-, and PI-functions is contained within a nonvolatile memory for storing bits of membership and parametric information in a plurality of address spaces. Based upon parametric and control signals, analog sensor data is digitized and converted into grade-of-membership data. In situ learn and recognition modes of operation are also provided.
Various instruments are used to create images of the Earth and other objects in the universe in a diverse set of wavelength bands with the aim of understanding natural phenomena. These instruments are sometimes built in a phased approach, with some measurement capabilities being added in later phases. In other cases, there may not be a planned increase in measurement capability, but technology may mature to the point that it offers new measurement capabilities that were not available before. In still other cases, detailed spectral measurements may be too costly to perform on a large sample. Thus, lower resolution instruments with lower associated cost may be used to take the majority of measurements. Higher resolution instruments, with a higher associated cost may be used to take only a small fraction of the measurements in a given area. Many applied science questions that are relevant to the remote sensing community need to be addressed by analyzing enormous amounts of data that were generated from instruments with disparate measurement capability. This paper addresses this problem by demonstrating methods to produce high accuracy estimates of spectra with an associated measure of uncertainty from data that is perhaps nonlinearly correlated with the spectra. In particular, we demonstrate multi-layer perceptrons (MLPs), Support Vector Machines (SVMs) with Radial Basis Function (RBF) kernels, and SVMs with Mixture Density Mercer Kernels (MDMK). We call this type of an estimator a Virtual Sensor because it predicts, with a measure of uncertainty, unmeasured spectral phenomena.
Ultramafic rocks are relatively rare at the Earth's surface but constitute the vast majority of the Earth by volume. Exposures of ultramafic bodies are therefore crucial for deducing many important processes that occur in the Earth's mantle. An important science question regarding the spatial distribution, abundance, and composition of mafic minerals in ultramafic bodies that can be examined with advanced sensor data is the melting process. When a lherzolite melts, clinopyroxene (cpx) melts first and therefore variations in the modal amount of cpx remaining in the mantle are a reflection of the amount of fractional melting that has occurred. Fe goes preferentially into the melt during melting but a 20 percent batch melting (i.e. closed system) acquires less Fe relative to 20 percent fractional melting (i.e. open system). Since the strength and wavelength of diagnostic absorptions is a strong function of Fe content, it is possible to make maps of the variation in Fe:Mg ratios which can be related to the general melting process. Accurate ground-truth information about local mineralogy provides internal calibration and consistency checks. Investigations using imaging spectrometer are very complementary to field studies because advanced sensor data can provide a synoptic view of modal mineralogy and chemical composition whereas field studies focus on detailed characterization of local areas. Two excellent exposures of ultramafic lithologies are being investigated with visible to mid-infrared imaging spectrometer data: the Ronda peridotite near Ronda, Spain and the Beni Bousera ophiolitic fragment in northern Morocco. Although separated by the Alboran Sea, these bodies are thought to be related and represent fertile sub-continental mantle. The Ronda peridotite is predominantly spinel lherzolite but grades into harzburgite and shows considerable variation in major and trace element compositions. Mafic layering and dykes (i.e. olivine gabbro) are also observed. This indicates some sections of the peridotite have experienced greater degrees of partial melting. The Beni Bousera peridotite also contains mafic layers and dykes and grades into harzburgite representing similar fundamental shifts in the bulk chemistry of this ultramafic body probably related to an episode of partial melting. The specific mode of emplacement of these bodies is controversial and important for understanding the tectonic evolution of this region. Our investigations are not necessarily designed to help resolve this controversy. Rather, these exposures provide excellent and unusual examples of fertile mantle which have undergone variable degrees of partial melting.
This work presents a computationally-ecient approach for damage determination that quanti es uncertainty in the provided diagnosis. Given strain sensor data that are polluted with measurement errors, Bayesian inference is used to estimate the location, size, and orientation of damage. This approach uses Bayes' Theorem to combine any prior knowledge an analyst may have about the nature of the damage with information provided implicitly by the strain sensor data to form a posterior probability distribution over possible damage states. The unknown damage parameters are then estimated based on samples drawn numerically from this distribution using a Markov Chain Monte Carlo (MCMC) sampling algorithm. Several modi cations are made to the traditional Bayesian inference approach to provide signi cant computational speedup. First, an ecient surrogate model is constructed using sparse grid interpolation to replace a costly nite element model that must otherwise be evaluated for each sample drawn with MCMC. Next, the standard Bayesian posterior distribution is modi ed using a weighted likelihood formulation, which is shown to improve the convergence of the sampling process. Finally, a robust MCMC algorithm, Delayed Rejection Adaptive Metropolis (DRAM), is adopted to sample the probability distribution more eciently. Numerical examples demonstrate that the proposed framework e ectively provides damage estimates with uncertainty quanti cation and can yield orders of magnitude speedup over standard Bayesian approaches.