Calibration of long wavelength Exotech model 20C spectroradiometer
Explore the source record for details and available documents.
Kumar, R.↗
Engineering topics
Kumar, R.
Publications and source records attributed to Kumar, R..
Separability of agricultural cover types in spectral channels and wavelength regions
This study was a continuation of a more complete evaluation of the spectral channels as well as wavelength regions - visible, near infrared, middle infrared, and thermal infrared - with respect to their estimated probability of correct classification Pc in discriminating agricultural cover types reported previously by Kumar and Silva (1977). Multispectral scanner data in twelve spectral channels in the wavelength range of 0.4-11.7 microns acquired in the middle of July for three flightlines were analyzed by applying automatic pattern recognition techniques. The same analysis was performed for the data acquired in the middle of August 1971, over the same three flightlines, to investigate the effect of time on the results. The effect of deletion of each spectral channel as well as each wavelength region on Pc is given. Values of Pc for all possible combinations of wavelength regions in the subsets of one to twelve spectral channels are also given. The overall values of Pc were found to be greater for the data of the middle of August than the data of the middle of July.
Kumar, R.↗
Calibration of long wavelength Exotech model 20C spectroradiometer
A brief description of the Exotech model 20C field spectroradiometer, which measures the spectral radiance of a target in the wavelength ranges from 0.37 to 2.5 microns (short wavelength unit), 2.8 to 5.6 microns, and 7.0 to 14 microns (long wavelength unit) is given. It is a rugged field instrument which uses four circular-variable-filters to get wavelength scan. Wavelength calibration of the long wavelength unit was done by use of the strong, sharp, and accurately known absorption bands of polystyrene, atmospheric carbon dioxide, and methyl cyclohexane (liquid) in the IR wavelength region. The spectral radiance calibration was done by recording spectral scans of the hot and the cold blackbody targets and assuming that spectral radiance varies linearly with the signal.
Kumar, R.↗
Effect of the atmosphere on the classification of LANDSAT data
The author has identified the following significant results. In conjunction with Turner's model for the correction of satellite data for atmospheric interference, the LOWTRAN-3 computer was used to calculate the atmospheric interference. Use of the program improved the contrast between different natural targets in the MSS LANDSAT data of Brasilia, Brazil. The classification accuracy of sugar canes was improved by about 9% in the multispectral data of Ribeirao Preto, Sao Paulo.
Dejesusparada, N.↗
Classification of areas using pixel-by-pixel and sample classifiers
There are no author-identified significant results in this report.
Dejesusparada, N.↗
Classification on accuracy of different options of the Image-100 system
There are no author-identified significant results in this report.
Dejesusparada, N.↗
Comparison of feature selection techniques for earth resources data
There are no author-identified significant results in this report.
Dejesusparada, N.↗
Calibration of longwavelength exotech model 20-C spectroradiometer
A brief description of the Exotech model 20-C field spectroradiometer which measures the spectral radiance of a target in the wavelength ranges 0.37 to 2.5 microns (short wavelength unit), 2.8 to 5.6 microns and 7.0 to 14 microns (long wavelength unit) is given. Wavelength calibration of long wavelength unit was done by knowing the strong, sharp and accurately known absorption bands of polystyrene, atmospheric carbon dioxide and methyl cyclohexane (liquid) in the infrared wavelength region. The spectral radiance calibration was done by recording spectral scans of the hot and the cold blackbodies and assuming that spectral radiance varies linearly with the signal.
Kumar, R.↗
Effects of systemic and non-systemic stresses on the thermal characteristics of corn
Experiments were conducted on corn plants using a calibrated spectroradiometer under field conditions in the indium antimonide channel (InSb, 2.8 to 5.6 mm) and the mercury cadmium telluride channel (HgCdTe, 7 to 14 mm). A ground cover experiment, an experiment on nonsystemic corn plants, and an experiment on systemic-stressed corn plants were included. The average spectral radiance temperature of corn plant populations was found (1) to be statistically significantly different for four healthy corn plant populations, (2) to increase with increased blight severity, and (3) to be statistically significantly different for varying rates of nitrogen applications.
Kumar, R.↗
Evaluation of wavelength groups for discrimination of agricultural cover types
Multispectral scanner data in twelve spectral channels, in the wavelength range 0.46 to 11.7 mm, acquired in July 1971 for three flightlines, were analyzed by applying automatic pattern recognition techniques. These twelve spectral channels were divided into four wavelength groups (W1, W2, W3 and W4), each consisting of three wavelength channels -- with respect to their estimated probability of correct classification (P sub c) in discriminating agricultural cover types. The same analysis was also done for the data acquired in August, to investigate the effect of time on these results. The effect of deletion of each of the wavelength groups on P sub C in the subsets of one to nine channels, is given. Values of P sub C for all possible combinations of wavelength groups, in the subsets of one to eleven channels are also given.
Kumar, R.↗
Evaluation of wavelength groups for discrimination of agricultural cover types
Multispectral scanner data in twelve spectral channels, in the wavelength range 0.46 to 11.7 microns, acquired in July, 1971, for three flightlines, were analyzed by applying automatic pattern recognition techniques. These twelve spectral channels were divided into four wavelength groups (W1, W2, W3 and W4), each consisting of three wavelength groups - with respect to their estimated probability of correct classification (Pc) - in discriminating agricultural cover types. The same analysis was also done for the data acquired in August, to investigate the effect of time on these results. The effect of deletion of each of the wavelength groups on Pc, in the subsets of one to nine channels, is given. Values of Pc for all possible combinations of wavelength groups, in the subsets of one to eleven channels, are also given.
Kumar, R.↗
Separability of agricultural cover types in spectral channels and wavelength regions
Spectral channels and wavelength regions (visible, near infrared, middle infrared and thermal infrared) were evaluated with respect to their estimated probability of correct classification (Pc) in discriminating agricultural cover types. Multispectral scanner data in twelve spectral channels in the wavelength range of 0.4 to 11.7 microns acquired in the middle of July for three flightlines, were analyzed by applying automatic pattern recognition techniques. The same analysis was performed for the data acquired in the middle of August, 1971, over the same three flightlines, to investigate the effect of time on the results. The effect of deletion of each spectral channel as well as each wavelength region on Pc is given. Values of Pc for all possible combinations of wavelength regions in the subsets of one to twelve spectral channels are also given. The overall values of Pc were found to be greater for the data of the middle of August than the data of the middle of July.
Kumar, R.↗
Radiometric correction of LANDSAT data
The author has identified the following significant results. The six independent sensors of the multispectral band scanner are supposed to be identical; however, in actual practice, they may have different gain settings and offset factors, which result in the effect known as stripping (black lines at regular intervals) of the imagery. A simple two parameter method to correct the gain settings and offset factors of each of the sensors with respect to one sensor, taken as reference, was developed. This method assumes: (1) the response of a detector varies linearly with the radiance of radiation received, and (2) the means, as well as the standard deviations, of a reasonably large number of pixels, in a given wavelength band, are equal for each of the detectors for the radiometrically corrected data.
Dejesusparada, N.↗
Effects of atmosphere, temperature and emittance on reflected and emitted energy
The effects of temperature and emittance on the relative magnitude of reflected energy and emitter energy from a target including atmospheric effects was studied. From the calculations of energy reflected and emitted from a target including atmospheric effects using LOWTRAN 3 programs for midlatitude summer model, the following conclusions were obtained: (1) At 3.5 micrometers q is considerably less than 1 except at high temperatures and for high emittance; (2) at 4 micrometers q is of the order of magnitude equal to 1 for most targets; and (3) at 4.6 micrometers, q is considerably greater than 1 at high temperatures and high emittance. In addition, incident atmospheric emission reflected from the target was found to be negligible except for targets having low temperature and low emittance.
Kumar, R.↗
Determination and error analysis of emittance and spectral emittance measurements by remote sensing
The author has identified the following significant results. From the theory of remote sensing of surface temperatures, an equation of the upper bound of absolute error of emittance was determined. It showed that the absolute error decreased with an increase in contact temperature, whereas, it increased with an increase in environmental integrated radiant flux density. Change in emittance had little effect on the absolute error. A plot of the difference between temperature and band radiance temperature vs. emittance was provided for the wavelength intervals: 4.5 to 5.5 microns, 8 to 13.5 microns, and 10.2 to 12.5 microns.
Dejesusparada, N.↗
Statistical separability and classification of land use classes using image-100
The author has identified the following significant results. The statistical separability of land use classes in the subsets of one to four spectral channels was investigated. Using ground observations and aerial photography, the MSS data of LANDSAT were analyzed with the Image-100. In the subsets of one to three spectral channels, channel 4, channel 4 & 7, and channels 4, 5, & 7 were found to be the best choices (ch.4 - 0.5 to 0.6 microns, ch. 5 - 0.6 to 0.7 microns, ch. 6 - 0.7 to 0.8 microns, and ch. 7 - 0.8 to 1.1 microns). For the single cell option of the Image-100, the errors of omission varied from 5% for the industrial class to 46% for the institutional class. The errors of commission varied from 11% for the commercial class to 39% for the industrial class. On the whole, the sample classifier gave considerably more accurate results compared to the single cell or multicell option.
Dejesusparada, N.↗
Evaluation of spectral channels and wavelength regions for separability of agricultural cover types
The author has identified the following significant results. Multispectral scanner data in twelve spectral channels in the wavelength range of 0.4 to 11.7 microns acquired in the middle of July for three flightlines were analyzed by applying automatic pattern recognition techniques. The same analysis was performed for the data acquired in mid August, over the same three flightlines, to investigate the effect of time on the results. The effect of deletion of each spectral channel, as well as each wavelength region on P sub c, is given. Values of P sub c for all possible combinations of wavelength regions in the subsets of one to twelve spectral channels are also given. The overall values of P sub c were found to be greater for the data of mid August than the data from mid July.
Dejesusparada, N.↗
Study of the Brazil and Falkland currents using their images of Nimbus 5 and oceanographic data in 1972 - 1973
The Western Edge of the Sub-tropical Convergence of the South-western Atlantic Ocean, called the Front, which is a thermal discontinuity between the Brazil and Falkland Currents, was studied utilizing the Temperature Humidity Infrared Radiometer (THIR) of Nimbus V in the 10.5 to 12.5 micrometers channel and historical oceanographic data. Some important results obtained are: the oceanographic Front could be detected from Nimbus THIR data; oceanographic charts showed that the transition zone where the Brazil and the Falkland Currents meet was the Front detected from satellite data; ocean current speeds calculated with THIR data were of the same order of magnitude as those calculated oceanographically; fisheries statistics for Pargo Roseo showed that the maximum catches were in September of 1973, in the period when the Front was observed most distinctly and clearly. The results showed the great potentiality of satellite data to study surface thermal structures, surface currents and oceanic fisheries.
Tseng, Y. C.↗