Evaluation of EO-1 hyperion data for agricultural applications
- PDF / 909,141 Bytes
- 12 Pages / 612 x 792 pts (letter) Page_size
- 19 Downloads / 220 Views
J. Indian Soc. Remote Sens. (September 2008) 36:255–266
RESEARCH ARTICLE
Evaluation of EO-1 Hyperion Data for Agricultural Applications Anshu Miglani . S. S. Ray . R. Pandey . J.S. Parihar
Received: 22 February 2006 / Accepted: 31 March 2008
Keywords Hyperion . Hyperspectral . Remote sensing . Atmospheric correction . Radiance . Principal component . Band selection . Band to band correlation
Abstract The present study was carried out to evaluate the satellite-based hyperspectral data available from Hyperion onboard EO-1 of NASA for agricultural applications. The study was carried out for Daurala block of Meerut district, using data of March 2005. The preliminary data analysis showed that there are 196 usable bands out of a total of 242 bands. Principal component (PC) analysis showed that about 99% of the information explained in 10 PCs. The atmospherically corrected reflectance, derived from satellite data had good agreement with the ground reflectance, observed using handheld spectroradiometer, with r2 ranging from 0.85 to 0.98. A. Miglani1 . S. S. Ray () . R. Pandey2 . J.S. Parihar ARG/RESIPA, Space Applications Center, ISRO, Ahmedabad - 380 015, India 1 A-145, GIS Development Pvt. Ltd., Sector 63, Noida - 201 301, India 2 Project Directorate for Cropping Systems Research, ICAR, Modipuram - 250 110, India
e-mail: [email protected]
A set of twenty most usable bands was selected by the criteria of maximum contribution to first five PCs and the band combinations with least inter-band correlations.
Introduction Hyperspectral remote sensing data provide spectroscopic information in relatively narrow contiguous spectral bands throughout visible, nearand short-wave infrared regions of the electromagnetic spectrum (Hong et al., 2002). The high spectral resolution reflectance spectrum in the region from 400–2500nm may be used to identify a large range of surface cover features that cannot be identified with broad band, low spectral resolution imaging system such as Landsat MSS, TM or SPOT (Hong et al., 2002; Goetz et al., 1985; Huete, 2003). Hyperspectral data can be used in discriminating crop varieties (Galvao et al., 2005), disease identification (Apan et al., 2004) and identification of stress within the crop.
123
256
In addition, hyperspectral data enable computation of narrowband indices using combined spectral bands related to biophysical parameters like LAI (Broge and Leblane, 2000; Pu et al., 2003; Meroni et al., 2004) and biochemical variables such as chlorophyll (Zarco-Tejada et al., 2004; Schaepman et al., 2005), water (Champagne et al., 2003) and nitrogen (Read et al., 2002). These parameters are robust indicators of the physiological and stress conditions that could potentially affect crop yield thus useful for precision agriculture purposes (ZarcoTejada et al., 2005). Furthermore, they provide an enhanced level of information for atmospheric correction (Datt et al., 2003; Galvao et al., 2005). With the availability of space-borne sensors having hyperspectral capabilities, a new era has open
Data Loading...
