The Hyperion Image Of Dehradun Analysis
seasonal streams and consists of varied landuse/landcover (LULC) classes like the crop fields, fallow land, urban, grassland, shrubs, and vegetation of mix type. The western part is covered with various vegetation features of Sal, Teak, and Bamboo. The southern part is covered with the urban built up area and some patches of vegetation. The urban area of the city is scattered and irregular. A seasonal river named Tons flows from
North East to South West direction.
5. Data Used
HYPERION is an EO-1 (Earth Observation-1) sensor which was developed under NASA’s new millennium program in November, 2000. The Hyperion image of Dehradun was acquired on 4th April, 2010. There are 224 bands with the spectral range of 355 to 2577 nm at 10 nm band width. …show more content…
2. The first step in the process was to preprocess the data to make it free from the errors that are implicit with most hyperspectral data. The preprocessing involves a rigorous radiometric and geometric correction. Spectrum of various classes has been extracted from the
HYPERION image after these corrections. Using a C program the spectral similarity measures have been computed for these spectra. In the last comparative analysis has been carried out using RSDPW.
7. Pre-processing of Hyperion data.
The pre-processing stage can be considered as the first step to work with the data sets available and acquire the valuable information from the same. The data sets that are available to the user initially are in the raw form and most of the datasets are not geo-referenced or geometrically calibrated to the ground. These datasets can also have errors due to atmospheric effects and need atmospheric corrections or radiometric calibrations. The users have data that contains lot of information and often user does not require the whole of it, therefore spatial and spectral subsetting of the data is required, which also helps in reducing the complexities and the processing …show more content…
Bad Band Removal
Hyperion level L1R data has 242 bands out of which only 198 are nonzero. Few were intentionally left unused
(Bands 1 to 7 and 225 to 242) and others fall in the overlap region of the two spectrometers (Bands 58 to 76).
Among the non zero bands, four band were still in the overlap region of the two spectrometers which are bands
56, 57 and 77, 78 out of which bands 77 and 78 were eliminated because of the higher noise levels present in those bands. Then there are water vapor absorption bands which needs to be eliminated and has been identified as bands 121 to 130 (1346 nm to 1467 nm), bands 165-182 (1800 to 1971 nm) and bands 220 (above 2356) and higher. Water vapors absorption bands absorb all the incident solar energy and can be easily identified visually.
The list of bands which were eliminated is given in the Table 2.
Apart from these bands some other bands were also eliminated containing the noisy data in those bands. After removing all the noisy bands from HYPERION data, set of 143 bands has been prepared for further
North East to South West direction.
5. Data Used
HYPERION is an EO-1 (Earth Observation-1) sensor which was developed under NASA’s new millennium program in November, 2000. The Hyperion image of Dehradun was acquired on 4th April, 2010. There are 224 bands with the spectral range of 355 to 2577 nm at 10 nm band width. …show more content…
2. The first step in the process was to preprocess the data to make it free from the errors that are implicit with most hyperspectral data. The preprocessing involves a rigorous radiometric and geometric correction. Spectrum of various classes has been extracted from the
HYPERION image after these corrections. Using a C program the spectral similarity measures have been computed for these spectra. In the last comparative analysis has been carried out using RSDPW.
7. Pre-processing of Hyperion data.
The pre-processing stage can be considered as the first step to work with the data sets available and acquire the valuable information from the same. The data sets that are available to the user initially are in the raw form and most of the datasets are not geo-referenced or geometrically calibrated to the ground. These datasets can also have errors due to atmospheric effects and need atmospheric corrections or radiometric calibrations. The users have data that contains lot of information and often user does not require the whole of it, therefore spatial and spectral subsetting of the data is required, which also helps in reducing the complexities and the processing …show more content…
Bad Band Removal
Hyperion level L1R data has 242 bands out of which only 198 are nonzero. Few were intentionally left unused
(Bands 1 to 7 and 225 to 242) and others fall in the overlap region of the two spectrometers (Bands 58 to 76).
Among the non zero bands, four band were still in the overlap region of the two spectrometers which are bands
56, 57 and 77, 78 out of which bands 77 and 78 were eliminated because of the higher noise levels present in those bands. Then there are water vapor absorption bands which needs to be eliminated and has been identified as bands 121 to 130 (1346 nm to 1467 nm), bands 165-182 (1800 to 1971 nm) and bands 220 (above 2356) and higher. Water vapors absorption bands absorb all the incident solar energy and can be easily identified visually.
The list of bands which were eliminated is given in the Table 2.
Apart from these bands some other bands were also eliminated containing the noisy data in those bands. After removing all the noisy bands from HYPERION data, set of 143 bands has been prepared for further