Search NASA⌕ Search

SEARCH · Search NASA

Results for “classifiers”

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 181 records · Page 10
All sourcesNASA NTRSDOE OSTI
NASA NTRS1973

Utilization of ERTS-1 data to monitor and classify eutrophication of inland lakes

The author has identified the following significant results. Significant findings are: (1) one-acre lakes and one-acre islands are detectable; (2)removal of atmospheric parameters derived from RPMI measurements show test lakes to have reflectances of 3.1 to 5.5% in band 4 and 0.3 to 2.3% in band 5; (3) failure to remove reflectance caused by atmosphere results in errors up to 500% in computing lake reflectance from ERTS-1 data; (4) in band 4, up to seven reflectance levels were observed in test lakes; (5) reflectance patterns have been displayed on a color-coded TV monitor and on computer-generated gray scales; (6) deep and shallow water can be separated by a trained photointerpreter and automatic machine processing, with estimates of water depth possible in some cases; (7) RPMI provides direct spectral signature measurements of lakes and lake features such as algal scums and floating plants; (8) a method is reported for obtaining lake color, as estimated by Forel-Ule standards, from ERTS-1 data; (9) a strong correlation between browner water color, diminishing water transparency; and (10) classifying lake eutrophication by observation of surface scums or macrophytes in shallow water seems straightforward.

Rogers, R. H.↗
NASA NTRS1974

Optimal design of an unsupervised adaptive classifier with unknown priors

An adaptive detection scheme for M hypotheses was analyzed. It was assumed that the probability density function under each hypothesis was known, and that the prior probabilities of the M hypotheses were unknown and sequentially estimated. Each observation vector was classified using the current estimate of the prior probabilities. Using a set of nonlinear transformations, and applying stochastic approximation theory, an optimally converging adaptive detection and estimation scheme was designed. The optimality of the scheme lies in the fact that convergence to the true prior probabilities is ensured, and that the asymptotic error variance is minimum, for the class of nonlinear transformations considered. An expression for the asymptotic mean square error variance of the scheme was also obtained.

Kazakos, D.↗
NASA NTRS1974

A new adaptive classifier using iterative filtering

To cope with signature variability, an algorithm has been defined which will adaptively classify remotely sensed data in the visible and near infrared band. The signal is divided into a space-dependent component and a target-dependent component. The target-dependent component is assumed fixed across the image for each target type. The space-dependent component is estimated iteratively by a weighted, least-squares algorithm. Included are the derivations of the sensor model and the two-dimensional, estimation algorithm.

Actkinson, A. L.↗
NASA NTRS1975

Adaptive statistical pattern classifiers for remotely sensed data

A technique for the adaptive estimation of nonstationary statistics necessary for Bayesian classification is developed. The basic approach to the adaptive estimation procedure consists of two steps: (1) an optimal stochastic approximation of the parameters of interest and (2) a projection of the parameters in time or position. A divergence criterion is developed to monitor algorithm performance. Comparative results of adaptive and nonadaptive classifier tests are presented for simulated four dimensional spectral scan data.

Gonzalez, R. C.↗
NASA NTRS1974

Implementation of an advanced table look-up classifier for large area land-use classification

Software employing Eppler's improved table look-up approach to pattern recognition has been developed, and results from this software are presented. The look-up table for each class is a computer representation of a hyperellipsoid in four dimensional space. During implementation of the software Eppler's look-up procedure was modified to include multiple ranges in order to accommodate hollow regions in the ellipsoids. In a typical ERTS classification run less than 6000 36-bit computer words were required to store tables for 24 classes. Classification results from the improved table look-up are identical with those produced by the conventional method, i.e., by calculation of the maximum likelihood decision rule at the moment of classification. With the new look-up approach an entire ERTS MSS frame can be classified into 24 classes in 1.3 hours, compared to 22.5 hours required by the conventional method. The new software is coded completely in FORTRAN to facilitate transfer to other digital computers.

Jones, C.↗
NASA NTRS1976

Selective radiant temperature mapping using a layered classifier

A method of measuring temperatures of selected ground-cover types using remotely sensed multispectral scanner data and a layered classification approach is described. A brief review of radiation theory is presented to show that for the wavelength bands and temperature ranges involved in remote sensing applications, a linear calibration function can be satisfactorily utilized. An example of the application of the layered classifier for temperature mapping of water is shown.

Bartolucci, L. A.↗
NASA NTRS1977

Tabular data base construction and analysis from thematic classified Landsat imagery of Portland, Oregon

A systematic verification of Landsat data classifications of the Portland, Oregon metropolitan area has been undertaken on the basis of census tract data. The degree of systematic misclassification due to the Bayesian classifier used to process the Landsat data was noted for the various suburban, industrialized and central business districts of the metropolitan area. The Landsat determinations of residential land use were employed to estimate the number of automobile trips generated in the region and to model air pollution hazards.

Bryant, N. A.↗
NASA NTRS1980

Automatically classifying Earth features from orbit

Solid state circuit classifies satellite imagery by spectral signature of vegetation, bare land, water, clouds, or snow. Circuit can be used to sort and separate specific imagery by signature so that only useful data is transmitted to earth. Device saves time and costs involved in manual separation of data.

Holstrom, R. L.↗
NASA NTRS1980

An automated approach to the design of decision tree classifiers

The classification of large dimensional data sets arising from the merging of remote sensing data with more traditional forms of ancillary data is considered. Decision tree classification, a popular approach to the problem, is characterized by the property that samples are subjected to a sequence of decision rules before they are assigned to a unique class. An automated technique for effective decision tree design which relies only on apriori statistics is presented. This procedure utilizes a set of two dimensional canonical transforms and Bayes table look-up decision rules. An optimal design at each node is derived based on the associated decision table. A procedure for computing the global probability of correct classfication is also provided. An example is given in which class statistics obtained from an actual LANDSAT scene are used as input to the program. The resulting decision tree design has an associated probability of correct classification of .76 compared to the theoretically optimum .79 probability of correct classification associated with a full dimensional Bayes classifier. Recommendations for future research are included.

Argentiero, P.↗
NASA NTRS1981

A comparison of two methods for classifying forestland

Two methods of developing land-cover classifications using Landsat multispectral data were compared. The first method, called P-1, uses a semi-automated approach to develop training statistics which characterize the land-cover types. The second, called multicluster blocks, depends more on analyst interaction to produce the training statistics used by the classifier. The results showed that P-1 performed as well as the multicluster-blocks approach on a mountainous study area in southwestern Colorado. These results may interest any resource discipline which has available to it ground-checked or photointerpreted information. P-1 can use this information directly to output a land-cover classification with little analyst interaction.

Nelson, R. F.↗
NASA NTRS1982

An automated approach to the design of decision tree classifiers

An automated technique is presented for designing effective decision tree classifiers predicated only on a priori class statistics. The procedure relies on linear feature extractions and Bayes table look-up decision rules. Associated error matrices are computed and utilized to provide an optimal design of the decision tree at each so-called 'node'. A by-product of this procedure is a simple algorithm for computing the global probability of correct classification assuming the statistical independence of the decision rules. Attention is given to a more precise definition of decision tree classification, the mathematical details on the technique for automated decision tree design, and an example of a simple application of the procedure using class statistics acquired from an actual Landsat scene.

Argentiero, P.↗
NASA NTRS1983

Method for classifying ceramic powder

Under the invented method, powder A of particles of less than 10 microns, and carrier powder B, whose average particle diameter is more than five times that of powder A, are premixed so that the powder is less than 40 wt.% of the total mixture, before classifying.

Takabe, K.↗
NASA NTRS1983

Histogram deconvolution - An aid to automated classifiers

It is shown that N-dimensional histograms are convolved by the addition of noise in the picture domain. Three methods are described which provide the ability to deconvolve such noise-affected histograms. The purpose of the deconvolution is to provide automated classifiers with a higher quality N-dimensional histogram from which to obtain classification statistics.

Lorre, J. J.↗
NASA NTRS1984

Segmentation of computer-classified Landsat multispectral scanner data into spatially-connected regions of elk habitat components

Segmentation of computer-classified Landsat multispectral data into spatially-connected regions of ground cover is described. Sorting and hash addressing, techniques commonly used for ordering and searching data records keyed by attributes, are the basis for the region identification and extraction. An example based on the use of spatial regions for evaluating elk habitat components is outlined.

Murray, R.↗
NASA NTRS1990

Using decision-tree classifier systems to extract knowledge from databases

One difficulty in applying artificial intelligence techniques to the solution of real world problems is that the development and maintenance of many AI systems, such as those used in diagnostics, require large amounts of human resources. At the same time, databases frequently exist which contain information about the process(es) of interest. Recently, efforts to reduce development and maintenance costs of AI systems have focused on using machine learning techniques to extract knowledge from existing databases. Research is described in the area of knowledge extraction using a class of machine learning techniques called decision-tree classifier systems. Results of this research suggest ways of performing knowledge extraction which may be applied in numerous situations. In addition, a measurement called the concept strength metric (CSM) is described which can be used to determine how well the resulting decision tree can differentiate between the concepts it has learned. The CSM can be used to determine whether or not additional knowledge needs to be extracted from the database. An experiment involving real world data is presented to illustrate the concepts described.

St.clair, D. C.↗
NASA NTRS1990

A failure diagnosis system based on a neural network classifier for the space shuttle main engine

A conceptual design of a model based failure detection and diagnosis system is developed for the space shuttle main engine. This design relies on the accurate and reliable identification of the parameters of the highly nonlinear and very complex engine. The design approach is presented in some detail and results for a failed valve are presented. These preliminary results verify that the developed parameter identification technique together with a neural network classifier can be used for this purpose.

Duyar, Ahmet↗
NASA NTRS1991

Hierarchical classifier design in high-dimensional, numerous class cases

As progress in new sensor technology continues, increasingly high spectral resolution sensors are being developed. These sensors give more detailed and complex data for each picture element and greatly increase the dimensionality of data over past systems. Three methods for designing a decision tree classifier are discussed; a top down approach, a bottom up approach, and a hybrid approach. Three feature extraction techniques are implemented. Canonical and extended canonical techniques are mainly dependent on the mean difference between two classes. An autocorrelation technique is dependent on the correlation differences. The mathematical relationship among sample size, dimensionality, and risk value is derived.

Kim, Byungyong↗
NASA NTRS1990

A failure diagnosis system based on a neural network classifier for the Space Shuttle Main Engine

A conceptual design of a model based failure deeteection and diagnosis system is deeveloped for the Space Shuttle Main Engine. This design relies on the accurate and reliable identification of the parameters of the highly nonlinear and very complex engine. The design approach is presented in some detail and results for a failed valve are presented. These preliminary results verify that the developed parameter identification technique, together with a neural network classifier, can be used for this purpose.

Duyar, Ahmet↗