SEARCH · Search NASA
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.
An approximation to a classification algorithm incorporating spatial context information in a general, statistical manner is presented which is computationally less intensive. Classifications that are nearly as accurate are produced.
The edited Guymon soil moisture data collected on August 2, 5, 14, 17, 1978 were grouped into four field cover types for statistical analysis. These are the bare, milo with rows parallel to field of view, milo with rows perpendicular to field of view and alfalfa cover groups. There are 37, 22, 24 and 14 observations respectively in each group for each sensor channel and each soil moisture layer. A subset of these data called the 'five cover set' (VEG5) limited the scatterometer data to the 15 deg look angle and was used to determine discriminant functions and combined group regressions.
The feasibility of dividing the 8-12 micrometer thermal infrared wavelength region into six spectral bands by an airborne line scanner system was investigated. By combining an existing scanner design with a 6 band spectrometer, a system for the remote sensing of Earth resources was developed. The elements in the spectrometer include an off axis reflective collimator, a reflective diffraction grating, a triplet germanium imaging lens, a photoconductive mercury cadmium telluride sensor array, and the mechanical assembly to hold these parts and maintain their optical alignment across a broad temperature range. The existing scanner design was modified to accept the new spectrometer and two field filling thermal reference sources.
Work done in combining spectral bands to reduce atmospheric effects on spectral signatures is described. The development of atmospheric models and their use with ground and aerial measurements in correcting spectral signatures is reviewed. An overview of studies of atmospheric effects on the accuracy of scene classification is provided.
A skin diagnosis system includes a scanning and optical arrangement whereby light reflected from each incremental area (pixel) of the skin is directed simultaneously to three separate light filters, e.g., IR, red, and green. As a result, the three devices simultaneously produce three signals which are directly related to the reflectance of light of different wavelengths from the corresponding pixel. These three signals for each pixel after processing are used as inputs to one or more output devices to produce a visual color display and/or a hard copy color print, for one usable as a diagnostic aid by a physician.
Early season changes in optical depth change brightness, primarily along the soil line; and during crop development, changes in optical depth change both greenness and brightness. Thus, the existence of haze in the imagery could cause an unsuspecting analyst to interpret the spectral appearance as indicating an episodal event when, in fact, haze was present. The techniques for converting LANDSAT-3 data to simulate LANDSAT-2 data are in error. The yellowness and none such computations are affected primarily. Yellowness appears well correlated to optical depth. Experimental evidence with variable background and variable optical depth is needed, however. The variance of picture elements within a spring wheat field is related to its equivalent in optical depth changes caused by haze. This establishes the sensitivity of channel 1 (greenness) pixels to changes in haze levels. The between field picture element means and variances were determined for the spring wheat fields. This shows the variability of channel data on two specific dates, emphasizing that crop development can be influenced by many factors. The atmospheric correction program ATCOR reduces segment data from LANDSAT acquisitions to a common haze level and improves the results of analysis.
Explore the source record for details and available documents.
A classification algorithm incorporating contextual information in a general, statistical manner is presented. Methods are investigated for obtaining adequate estimates of the context distribution (a statistical characterization of context) upon which the classification algorithm depends. Finally, a method of estimating optimal algorithm parameters prior to performing preliminary classifications is explored.
A solar radiation model is described which is able to calculate, for given atmospheric and topographic input conditions, the spectral distributions of the direct and diffuse components of the total irradiance incident upon a target surface. The irradiance includes both upwelling radiation backscattered toward the earth's surface and radiation reflected toward the target from other terrain surfaces. Spectral reflectance data for two dissimilar snow types are fed into the model, along with a geometric description of the topography of the snow covered terrain. These data permit the calculation of the exitance of the target over the wavelength limits of a given sensor and, hence, the apparent contrast between the two snow types for the given input conditions. The results show that the contrast between the two snow types, all other factors being constant, is greatest in the near infrared wavelengths.
The relationship existing between Guanabara Bay water quality ground truth parameters and LANDSAT MSS video data was investigated. The parameters considered were: chorophyll content, water transparency usng the Secchi disk, salinity, and dissolved ammonia. Data from two overflights was used, and methods of processing digital data were compared. Linear and nonlinear regression analyses were utilized, comparing original data with processed data by using the correlation coefficient and the estimation mean error. It was determined that better quality data are obtained by using radiometric correction programs with a physical basis, contrast ratio, and normalization. Incidental locations of floating vegetation, changes in bottom depth, oil slicks, and ships at anchor were made.
An algorithm is proposed which predicts the optimal features at every node in a binary tree procedure. The algorithm estimates the probability of error by approximating the area under the likelihood ratio function for two classes and taking into account the number of training samples used in estimating each of these two classes. Some results on feature selection techniques, particularly in the presence of a very limited set of training samples, are presented. Results comparing probabilities of error predicted by the proposed algorithm as a function of dimensionality as compared to experimental observations are shown for aircraft and LANDSAT data. Results are obtained for both real and simulated data. Finally, two binary tree examples which use the algorithm are presented to illustrate the usefulness of the procedure.
The improvement in land cover classification achieved by merging microwave data with Landsat MSS data is examined. To produce a merged data set for analysis and comparison, a registration procedure by which a set of Seasat SAR digital data was merged with the MSS data is described. The Landsat MSS data and the merged Landsat/Seasat data sets were processed using conventional multichannel spectral pattern recognition techniques. An analysis of the classified data sets indicates that while Landsat data delineate different forest types (i.e., deciduous/coniferous) and allow some species separation, SAR data provide additional information related to plant canopy configuration and vegetation density as associated with varying water regimes, and therefore allow for further subdivision in the classification of forested wetlands of the coastal region of the southern United States.
An instrument concept that uses solid-state array imaging has been developed for a future land observing system. The design concept is responsive to a variety of use needs and provides improved capabilities over the planned Landsat Thematic Mapper. A comparison of the differing approaches to the instrument design was made, resulting in the selection of a concept which uses a spectrograph coupled to a line-array imager to provide simultaneous spatial and spectral resolution. The design provides an inherent solution to the problem of achieving precise registration among the spectral bands. Data processing on the focal plane is used to select the spectral bands and their band widths. Onboard capabilities include radiometric correction, selection of instantaneous field-of-view and swath width, and data compression.
Characteristics of tropospheric desert aerosols are derived by comparing nadir spectral reflectivities computed from the radiative transfer models with reflectivities measured from Landsat. Over the ocean, reflectivities are compared, but over land the comparison is carried out by determining the ratios of the nadir reflectivity of the surface-atmosphere system over heavy aerosol concentration to the reflectivity of the underlying surface. This remote sensing technique is found to be a sensitive approach for measuring n sub 2, the imaginary part of the refractive index. The desert aerosols under study, in the Iran and Pakistan area, are essentially pure scatterers, inasmuch as an n sub 2 value of 0.001 + or - 0.001 was determined for each of the four Landsat spectral bands, that is, for a spectral interval from 0.5 to 1.1 microns.
Recent investigations have demonstrated the effectiveness of a contextual classifier that combines spatial and spectral information employing a general statistical approach. This statistical classification algorithm exploits the tendency of certain ground-cover classes to occur more frequently in some spatial contexts than in others. Indeed, a key input to this algorithm is a statistical characterization of the context: the context distribution. Here a discussion is given of an unbiased estimator of the context distribution which, besides having the advantage of statistical unbiasedness, has the additional advantage over other estimation techniques of being amenable to an adaptive implementation in which the context distribution estimate varies according to local contextual information. Results from applying the unbiased estimator to the contextual classification of three real Landsat data sets are presented and contrasted with results from noncontextual classifications and from contextual classifications utilizing other context distribution estimation techniques.
Explore the source record for details and available documents.
Spectrometer scan head redesigned to accommodate larger spectrometer and two blackbody reference sources while remaining within space limitations of its aircraft mounting.
A computational model of the deterministic and stochastic process of remote sensing has been developed based upon the results of the investigations presented. The model is used in studying concepts for improving worldwide environment and resource monitoring. A review of various atmospheric radiative transfer models is presented as well as details of the selected model. Functional forms for spectral diffuse reflectance with variability introduced are also presented. A cloud detection algorithm and the stochastic nature of remote sensing data with its implications are considered.