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 289 records · Page 16
All sourcesNASA NTRSDOE OSTI
NASA NTRS1970

Classifying forest and nonforest land on space photographs

Although the research reported is in its preliminary stages, results show that: (1) infrared color film is the best single multiband sensor available; (2) there is a good possibility that forest can be separated from all nonforest land uses by microimage evaluation techniques on IR color film coupled with B/W infrared and panchromatic films; and (3) discrimination of forest and nonforest classes is possible by either of two methods: interpreters with appropriate viewing and mapping instruments, or programmable automatic scanning microdensitometers and automatic data processing.

Aldrich, R. C.↗
NASA NTRS1971

Layered recognition networks that pre-process, classify, and describe

A brief overview is presented of six types of pattern recognition programs that: (1) preprocess, then characterize; (2) preprocess and characterize together; (3) preprocess and characterize into a recognition cone; (4) describe as well as name; (5) compose interrelated descriptions; and (6) converse. A computer program (of types 3 through 6) is presented that transforms and characterizes the input scene through the successive layers of a recognition cone, and then engages in a stylized conversation to describe the scene.

Uhr, L.↗
NASA NTRS1972

Layered 'recognition cone' networks that pre-process, classify, and describe.

Discussion of pattern recognition programs for input data preprocessing with simultaneous or subsequent characterization, or characterization into a 'recognition cone,' or description and naming, interrelated descriptions, and conversion. A computer program is described that transforms and characterizes the input through the successive layers of a recognition cone. The program can choose and put forth names of parts of the input scene. It combines pieces of a description into interrelated wholes by using n-tuple characterizers and conducts a simple and stylized conversation about what it has seen. The technique of combining recognition cones with preprocessing transformations and characterizations is expected to contribute to technology in this field.

Uhr, L.↗
NASA NTRS1972

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

The author has identified the following significant results. Channel 7 is highly useful in surveying surface water in lakes, rivers, and large marshes. It is likely to miss detection of elliptically shaped bodies of 4 acres or less. Further, it is possible that the bodies are distorted and displaced because of lack of correction for sensor response time. These errors might not be critical because:(1) location accuracy is not essential to a surface water survey; and (2) an obviously distorted image is often not in error in excess of 5% The finding that Orchard Lake and other lakes in Oakland County have different densities in channels 4, 5, 6, and 7 is important because it implies that the lake wide water color average is different in the separate channels. Channels 6 and 7 were constant in tonal quality among all the lakes while channels 4 and 5 varied from lake to lake and in various parts of Orchard Lake. These findings are significant because it means that small inland lake color differences are recorded by the MSS even on a cloudy or hazy day. It also confirms that ERTS-1 is performing well enough to be used for correlation to ground truth and aircraft underflights.

Chase, P. E.↗
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) circular lakes of 7.5 acres and greater reach full density; (3) long channels 100 ft wide are detectable; (4) orientation of lakes is independent of scan direction; (5) lake features are observable in enlargements of CCT imagery produced in the Bendix Earth Resources Data Center; and (6) a decision surface water outline map is presented that was produced from ERTS-1 CCT. A water color literature review, baseline water quality data of the test lakes, and a discussion of geometric corrections of the CCT decision water surface outline map are also presented.

Chase, P. E.↗
NASA NTRS1970

The use of space and high altitude aerial photography to classify forest land and to detect forest disturbances

In October 1969, an investigation was begun near Atlanta, Georgia, to explore the possibilities of developing predictors for forest land and stand condition classifications using space photography. It has been found that forest area can be predicted with reasonable accuracy on space photographs using ocular techniques. Infrared color film is the best single multiband sensor for this purpose. Using the Apollo 9 infrared color photographs taken in March 1969 photointerpreters were able to predict forest area for small units consistently within 5 to 10 percent of ground truth. Approximately 5,000 density data points were recorded for 14 scan lines selected at random from five study blocks. The mean densities and standard deviations were computed for 13 separate land use classes. The results indicate that forest area cannot be separated from other land uses with a high degree of accuracy using optical film density alone. If, however, densities derived by introducing red, green, and blue cutoff filters in the optical system of the microdensitometer are combined with their differences and their ratios in regression analysis techniques, there is a good possibility of discriminating forest from all other classes.

Aldrich, R. C.↗
NASA NTRS1973

The use of the temporal dimension in classifying and mapping ERTS-1 MSS data

Multispectral data from two ERTS-1 scenes of the same central Pennsylvania area were brought into registration by translation and then merged. The two scenes were viewed on different dates, but from adjacent ground tracks, as frequent cloud cover in Pennsylvania made it impossible to choose two scenes from the same track. Targets selected to be mapped included river water, railroad yards, creeks, urban areas, industrial areas, and vegetation. Equivalent training areas were chosen from each of the original scenes and from the merged data. Classification maps were produced for each, and a comparison was made. Scene brightness was found to have the most important effect on classification differences.

Borden, F. Y.↗
NASA NTRS1973

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

A technique is being developed for use of ERTS in estimating and monitoring trophic levels of inland lakes. Preliminary findings are that Michigan lakes and ponds of one acre or more are resolvable in bands 5, 6 and 7 of NASA MSS imagery under fair conditions (haze and 70% cloud cover). In processed imagery (CCT) smaller features, including water color patterns, are evident within some lakes of 40 acres or more. Image distortion of lake size, shape, orientation, etc. is minimal; discrimination of lakes and ponds from various wetlands is good. Subsequent ERTS and aircraft imagery will be correlated with detailed ground truth of water color and quality in eutrophic test lakes.

Chase, P. E.↗
NASA NTRS1973

Classifying unresolved objects from simulated space data.

A multispectral scanner data set gathered at a flight altitude of 10,000 ft. over an agricultural area was modified to simulate the spatial resolution of the spacecraft scanners. Signatures were obtained for several major crops and their proportions were estimated over a large area. For each crop, a map was generated to show its approximate proportion in each resolution element, and hence its distribution over the area of interest. A statistical criterion was developed to identify data points that may not represent a mixture of the specified crops. This allows for great reduction in the effect of unknown or alien objects on the estimated proportions. This criterion can be used to locate special features, such as roads or farm houses. Preliminary analysis indicates a high level of consistency between estimated proportions and available ground truth. Large concentrations of major crops show up especially well on the maps.

Nalepka, R. F.↗
NASA NTRS1975

Classifying and monitoring water quality by use of satellite imagery

A technique in which LANDSAT measurements from very clear lakes are subtracted from measurements from other lakes in order to remove atmospheric and surface noise effects to obtain a residual signal dependent only on the material suspended in the water is described. This residual signal is used by the Multispectral Data Analysis System as a basis for producing color categorized imagery showing lakes by type and concentration of suspended material. Several hundred lakes in the Madison and Spooner, Wisconsin area were categorized for tannin or non-tannin waters and for the degree of algae, silt, weeds, and bottom effects.

Scherz, J. P.↗
NASA NTRS1976

Development of a methodology for classifying software errors

A mathematical formalization of the intuition behind classification of software errors is devised and then extended to a classification discipline: Every classification scheme should have an easily discernible mathematical structure and certain properties of the scheme should be decidable (although whether or not these properties hold is relative to the intended use of the scheme). Classification of errors then becomes an iterative process of generalization from actual errors to terms defining the errors together with adjustment of definitions according to the classification discipline. Alternatively, whenever possible, small scale models may be built to give more substance to the definitions. The classification discipline and the difficulties of definition are illustrated by examples of classification schemes from the literature and a new study of observed errors in published papers of programming methodologies.

Gerhart, S. L.↗