SEARCH · Engineering Papers

Results for “Image Classification”

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.

SourceResearch connectionFrom yearThrough year

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

At least 577 records · Page 32

All sources NASA NTRS DOE OSTI

NASA NTRS1994

Effect of radiance-to-reflectance transformation and atmosphere removal on maximum likelihood classification accuracy of high-dimensional remote sensing data

Many analysis algorithms for high-dimensional remote sensing data require that the remotely sensed radiance spectra be transformed to approximate reflectance to allow comparison with a library of laboratory reflectance spectra. In maximum likelihood classification, however, the remotely sensed spectra are compared to training samples, thus a transformation to reflectance may or may not be helpful. The effect of several radiance-to-reflectance transformations on maximum likelihood classification accuracy is investigated in this paper. We show that the empirical line approach, LOWTRAN7, flat-field correction, single spectrum method, and internal average reflectance are all non-singular affine transformations, and that non-singular affine transformations have no effect on discriminant analysis feature extraction and maximum likelihood classification accuracy. (An affine transformation is a linear transformation with an optional offset.) Since the Atmosphere Removal Program (ATREM) and the log residue method are not affine transformations, experiments with Airborne Visible/Infrared Imaging Spectrometer (AVIRIS) data were conducted to determine the effect of these transformations on maximum likelihood classification accuracy. The average classification accuracy of the data transformed by ATREM and the log residue method was slightly less than the accuracy of the original radiance data. Since the radiance-to-reflectance transformations allow direct comparison of remotely sensed spectra with laboratory reflectance spectra, they can be quite useful in labeling the training samples required by maximum likelihood classification, but these transformations have only a slight effect or no effect at all on discriminant analysis and maximum likelihood classification accuracy.

Hoffbeck, Joseph P.↗

NASA NTRS1983

Analysis of multispectral data using an unsupervised classification technique: Application to VAS

A statistical classification method based on clustering of multidimensional histograms was applied to several channels of the VAS multispectral imagery. The method automatically discriminates and classifies atmospheric ground features such as cloud types, atmospheric moisture patterns, ocean, or ground. Such a clustering method has the advantage of forming natural data groupings, without a priori classification. Clusters are not limited by straight lines or plane surfaces as is the case in threshold methods. The method was applied to simultaneous full resolution images from channels 8 (11.2 micron), 10 (6.7 micron), and 12 (3.9 micron). Twenty image segments of 64 by 64, 12 image segments of 128 by 128, and 4 image segments of 254 by 254 picture elements were analyzed. In addition, normal VISSR mode images at 1800, 1830, and 2000 GMT were used to identify the classes. The gray levels measured along a scan line and the result of the classification scheme (dashed curves) for the three channels investigated are shown. Each point of the image is affected to a class. Each class is identified by a center of gravity that is represented by a vector in the three dimensional space of gray levels.

Szejwach, G.↗

NASA NTRS1991

Data space volumes and classification optimization of SPOT and Landsat TM data

In order to compare the data space volume of SPOT XS and Landsat TM images, three data sets, i.e., a wetlands/agricultural data set, an agricultural data set, and a forest data set, are examined. The comparisons are made for the same geographic area. The data space volumes for Landsat TM (2, 3, and 4) are found to be 70 to 100 percent larger than the volumes for the SPOT XS images. It is suggested that the additional midinfrared bands contribute to the difference in data space volumes between Landsat TM and SPOT XS. The data space volumes for Landsat TM bands 3, 4, and 5 are more than an order of magnitude greater than the volumes for the three band SPOT XS data sets. The volumes of the six-band Landsat TM images are four orders of magnitude greater than the SPOT XS. The analysis of the data space volumes is used to optimize the computation time and minimize the storage requirements of a maximum-likelihood classification based on a look-up table.

Ahearn, Sean C.↗

NASA NTRS1981

Application of SEASAT-1 Synthetic Aperture Radar (SAR) data to enhance and detect geological lineaments and to assist LANDSAT landcover classification mapping

Digital SEASAT-1 synthetic aperture radar (SAR) data were used to enhance linear features to extract geologically significant lineaments in the Appalachian region. Comparison of Lineaments thus mapped with an existing lineament map based on LANDSAT MSS images shows that appropriately processed SEASAT-1 SAR data can significantly improve the detection of lineaments. Merge MSS and SAR data sets were more useful fo lineament detection and landcover classification than LANDSAT or SEASAT data alone. About 20 percent of the lineaments plotted from the SEASAT SAR image did not appear on the LANDSAT image. About 6 percent of minor lineaments or parts of lineaments present in the LANDSAT map were missing from the SEASAT map. Improvement in the landcover classification (acreage and spatial estimation accuracy) was attained by using MSS-SAR merged data. The aerial estimation of residential/built-up and forest categories was improved. Accuracy in estimating the agricultural and water categories was slightly reduced.

Sekhon, R.↗

NASA NTRS1994

Unsupervised Segmentation Of Polarimetric SAR Data

Method of unsupervised segmentation of polarimetric synthetic-aperture-radar (SAR) image data into classes involves selection of classes on basis of multidimensional fuzzy clustering of logarithms of parameters of polarimetric covariance matrix. Data in each class represent parts of image wherein polarimetric SAR backscattering characteristics of terrain regarded as homogeneous. Desirable to have each class represent type of terrain, sea ice, or ocean surface distinguishable from other types via backscattering characteristics. Unsupervised classification does not require training areas, is nearly automated computerized process, and provides nonsubjective selection of image classes naturally well separated by radar.

Rignot, Eric J.↗

NASA NTRS1984

An application of expert systems technology to remotely sensed image analysis

In October of 1983, a project was initiated at NASA/Ames Research Center to design a prototype expert system capable of producing a preliminary land cover classification from an unsupervised classification of Landsat Multispectral Scanner (MSS) imagery and associated collateral data. This paper outlines the problem of determining land cover classes from an unsupervised classification, presents a brief overview of expert systems, and continues with a description of the design and current state of implementation of NASA/Ames' prototype Landsat MSS image analysis expert system.

Erickson, W. K.↗

NASA NTRS1975

A comparative interregional analysis of selected data from LANDSAT-1 and EREP for the inventory and monitoring of natural ecosystems

Comparative statistics were presented on the capability of LANDSAT-1 and three of the Skylab remote sensing systems (S-190A, S-190B, S-192) for the recognition and inventory of analogous natural vegetations and landscape features important in resource allocation and management. Two analogous regions presenting vegetational zonation from salt desert to alpine conditions above the timberline were observed, emphasizing the visual interpretation mode in the investigation. An hierarchical legend system was used as the basic classification of all land surface features. Comparative tests were run on image identifiability with the different sensor systems, and mapping and interpretation tests were made both in monocular and stereo interpretation with all systems except the S-192. Significant advantage was found in the use of stereo from space when image analysis is by visual or visual-machine-aided interactive systems. Some cost factors in mapping from space are identified. The various image types are compared and an operational system is postulated.

Poulton, C. E.↗

NASA NTRS1977

Satellite land use acquisition and applications to hydrologic planning models

A developing operational procedure for use by the Corps of Engineers in the acquisition of land use information for hydrologic planning purposes was described. The operational conditions preclude the use of dedicated, interactive image processing facilities. Given the constraints, an approach to land use classification based on clustering seems promising and was explored in detail. The procedure is outlined and examples of application to two watersheds given.

Algazi, V. R.↗

NASA NTRS1986

The stratigraphy of Mars

A global stratigraphy of Mars was developed from a global geologic map series derived from Viking images; the stratigraphy is composed of three maps. A new chronostratigraphic classification system which consists of lower, middle, and upper Noachian, Hesperian, and Amazonian systems is described. The crater-density boundaries of the chronostratigraphic units and the absolute ages of the Martian epochs aer estimated. The relative ages of major geologic units and featues are calculated and analyzed. The geologic history of Mars is summarized on the maps in terms of epochs.

Tanaka, Kenneth L.↗

NASA NTRS1987

Gaussian maximum likelihood and contextual classification algorithms for multicrop classification

The paper reviews some of the ways in which context has been handled in the remote-sensing literature, and additional possibilities are introduced. The problem of computing exhaustive and normalized class-membership probabilities from the likelihoods provided by the Gaussian maximum likelihood classifier (to be used as initial probability estimates to start relaxation) is discussed. An efficient implementation of probabilistic relaxation is proposed, suiting the needs of actual remote-sensing applications. A modified fuzzy-relaxation algorithm using generalized operations between fuzzy sets is presented. Combined use of the two relaxation algorithms is proposed to exploit context in multispectral classification of remotely sensed data. Results on both one artificially created image and one MSS data set are reported.

Di Zenzo, Silvano↗

NASA NTRS1990

Opening and closing of sea ice leads - Digital measurements from synthetic aperture radar

An algorithm that uses two sequential SAR digital images to measure the opening and closing of sea ice leads is introduced. The methods used to analyze the SAR images is described, including the mesh of tie points and the classification of leads and ice. The results of the anaylsis are compared with independent manual measurements, showing that the cells covering each lead are best interpreted as opening or closing in a group, rather than individually. Also, consideration is given to an automated algorithm for grouping cells, the possibility of simplifying the method, and the relationship between the opening and closing measurements and the theory of their parametric relation to mean deformation.

Fily, M.↗

NASA NTRS1994

The morphology of faint galaxies in Medium Deep Survey images using WFPC2

First results from Hubble Space Telescope (HST) Medium Deep Survey images taken with Wide Field/Planetary Camera-2 (WFPC2) demonstrate that galaxy classifications can be reliably performed to magnitudes I814 approximately less than 22.0 in the F815W band. Published spectroscopic surveys to this depth indicate a mean redshift of bar-z approximately 0.5. We have classified over 200 galaxies in nine WFPC2 fields according to a basic morphological scheme. The majority of these faint galaxies appear to be similar to regular Hubble-sequence examples observed at low redshift. To the precision of our classification scheme, the relative proportion of spheroidal and disk systems of normal appearance is as expected from nearby samples, indicating that the bulk of the local galaxy population was in place at half the Hubble time. However, the most intriguing result is the relatively high proportion (approximately 40%) of objects which are in some way anomalous, and which may be of relevance in understanding the origin of the familiar excess population of faint galaxies established by others. These diverse objects include apparently interacting pairs whose multiple structure is only revealed with HST's angular resolution, galaxies with superluminous star-forming regions, diffuse low surface brightness galaxies of various forms, and compact galaxies. These anomalous galaxies contribute a substantial fraction of the excess counts at our limiting magnitude, and may provide insights into the 'faint blue galaxy' problem.

Griffiths, R. E.↗

NASA NTRS2002

Wire Detection Algorithms for Navigation

In this research we addressed the problem of obstacle detection for low altitude rotorcraft flight. In particular, the problem of detecting thin wires in the presence of image clutter and noise was studied. Wires present a serious hazard to rotorcrafts. Since they are very thin, their detection early enough so that the pilot has enough time to take evasive action is difficult, as their images can be less than one or two pixels wide. Two approaches were explored for this purpose. The first approach involved a technique for sub-pixel edge detection and subsequent post processing, in order to reduce the false alarms. After reviewing the line detection literature, an algorithm for sub-pixel edge detection proposed by Steger was identified as having good potential to solve the considered task. The algorithm was tested using a set of images synthetically generated by combining real outdoor images with computer generated wire images. The performance of the algorithm was evaluated both, at the pixel and the wire levels. It was observed that the algorithm performs well, provided that the wires are not too thin (or distant) and that some post processing is performed to remove false alarms due to clutter. The second approach involved the use of an example-based learning scheme namely, Support Vector Machines. The purpose of this approach was to explore the feasibility of an example-based learning based approach for the task of detecting wires from their images. Support Vector Machines (SVMs) have emerged as a promising pattern classification tool and have been used in various applications. It was found that this approach is not suitable for very thin wires and of course, not suitable at all for sub-pixel thick wires. High dimensionality of the data as such does not present a major problem for SVMs. However it is desirable to have a large number of training examples especially for high dimensional data. The main difficulty in using SVMs (or any other example-based learning method) is the need for a very good set of positive and negative examples since the performance depends on the quality of the training set.

Kasturi, Rangachar↗

NASA NTRS2023

Capitol Reef Ecological Conservation: Mapping Vegetation Functional Groups to Inform Invasive Vegetation Management, Ecological Conservation and Restoration in Capitol Reef National Park

Invasive exotic plant (IEP) species have been found within the park boundaries of Capitol Reef National Park (CARE) in Utah. Currently, remotely sensed datasets such as the Rangeland Analysis Platform (RAP) from the United States Department of Agriculture (USDA) have been used to investigate IEP species within the park, but validation of the national RAP program is necessary for informing decisions at a local scale. CARE seeks a remote monitoring solution that can precisely target managerial efforts within the park’s challenging terrain and hard-to-reach locations. To fulfill this objective, we harnessed Landsat 8 Operational Land Imager (OLI) imagery and leveraged Random Forest (RF) modeling to generate classification maps characterizing vegetation functional groups for 2013 and 2022 within the park. Subsequently, the Land Change Modeler (LCM) in Idrisi TerrSet facilitated the production of a predicted classification map for 2033. The team also devised an annual grass probability map to accentuate areas impacted by exotic grasses. A comparative assessment between the RF classification map and the RAP map for 2022 revealed an overall agreement of 47.41%, with disparities primarily arising from differences in bare soil and shrub areas. Significantly, the 2022 RF-generated classification map showcased an impressive overall accuracy of 92.17%. In short, the probability map, the land cover change detection spanning 2013 to 2022, and the forecasting of observed trends into the future aids in the evaluation of invasive plant impacts and facilitation of CARE’s preparedness for potential ecological disturbances. Notably, in comparison to the RAP, the RF classification method generates functional group maps that are more representative of the study area.

Vanchy Li↗

NASA NTRS1977

Necessity to adapt land use and land cover classification systems to readily accept radar data

A hierarchial, four level, standardized system for classifying land use/land cover primarily from remote-sensor data (USGS system) is described. The USGS system was developed for nonmicrowave imaging sensors such as camera systems and line scanners. The USGS system is not compatible with the land use/land cover classifications at different levels that can be made from radar imagery, and particularly from synthetic-aperture radar (SAR) imagery. The use of radar imagery for classifying land use/land cover at different levels is discussed, and a possible revision of the USGS system to more readily accept land use/land cover classifications from radar imagery is proposed.

Drake, B.↗

NASA NTRS2000

From Stars to Super-Planets: The Low-Mass IMF in the Young Cluster IC348

We investigate the low-mass population of the young cluster IC348 down to the deuterium-burning limit, a fiducial boundary between brown dwarf and planetary mass objects, using a new and innovative method for the spectral classification of late-type objects. Using photometric indices, constructed from HST/NICMOS narrow-band imaging, that measure the strength of the 1.9 micron water band, we determine the spectral type and reddening for every M-type star in the field, thereby separating cluster members from the interloper population. Due to the efficiency of our spectral classification technique, our study is complete from approximately 0.7 solar mass to 0.015 solar mass. The mass function derived for the cluster in this interval, dN/d log M alpha M(sup 0.5), is similar to that obtained for the Pleiades, but appears significantly more abundant in brown dwarfs than the mass function for companions to nearby sun-like stars. This provides compelling observational evidence for different formation and evolutionary histories for substellar objects formed in isolation vs. as companions. Because our determination of the IMF is complete to very low masses, we can place interesting constraints on the role of physical processes such as fragmentation in the star and planet formation process and the fraction of dark matter in the Galactic halo that resides in substellar objects.

Najita, Joan R.↗

NASA NTRS1986

The effect of Thematic Mapper spectral properties on land cover mapping for hydrologic modeling

The accuracy of unsupervised land-cover classification from all seven Landsat TM bands and from six combinations of three or four bands is evaluated using images of the Clinton River Basin, a suburban watershed near Detroit. Data from aerial TMS photography, USGS topographic maps, and ground surveys are employed to determine the classification accuracy. The mapping accuracy of all seven bands is found to be significantly better (6 percent overall, 12 percent for residential areas, and 13 percent for commercial districts) than that with bands 2, 3, and 4; but almost the same accuracy is obtained by including at least one band from each major spectral region (visible, NIR, and mid-IR).

Gervin, J. C.↗

NASA NTRS1995

The composite classification problem in optical information processing

Optical pattern recognition allows objects to be recognized from their images and permits their positional parameters to be estimated accurately in real time. The guiding principle behind optical pattern recognition is that a lens focusing a beam of coherent light modulated with an image produces the two-dimensinal Fourier transform of that image. When the resulting output is further transformed by the matched filter corresponding to the original image, one obtains the autocorrelation function of the original image, which has a peak at the origin. Such a device is called an optical correlator and may be used to recognize the locate the image for which it is designed. (From a practical perspective, an approximation to the matched filter must be used since the spatial light modulator (SLM) on which the filter is implemented usually does not allow one to independently control both the magnitude and phase of the filter.) Generally, one is not just concerned with recognizing a single image but is interested in recognizing a variety of rotated and scaled views of a particular image. In order to recognize these different views using an optical correlator, one may select a subset of these views (whose elements are called training images) and then use a composite filter that is designed to produce a correlation peak for each training image. Presumably, these peaks should be sharp and easily distinguishable from the surrounding correlation plane values. In this report we consider two areas of research regarding composite optical correlators. First, we consider the question of how best to choose the training images that are used to design the composite filter. With regard to quantity, the number of training images should be large enough to adequately represent all possible views of the targeted object yet small enough to ensure that the resolution of the filter is not exhausted. As for the images themselves, they should be distinct enough to avoid numerical difficulties yet similar enough to avoid gaps in which certain views of the target will be unrecognized. One method that we introduce to study this problem is called probing and involves the creation of the artificial imagery. The second problem we consider involves the clasification of the composite filter's correlation plane data. In particular, we would like to determine not only whether or not we are viewing a training image, but, in the former case, we would like to determine which training image is being viewed. This second problem is investigated using traditional M-ary hypothesis testing techniques.

Hall, Eric B.↗