International Journal of Applied Earth Observation and Geoinformation
Analysis of spectral absorption features in hyperspectral imagery
Introduction
When light interacts with a mineral or rock, light of certain wavelengths is preferentially absorbed while at other wavelengths it is transmitted in the substance. Reflectance, is defined as the ratio of the intensity of light reflected from a sample to the intensity of the light incident on it. Electronic transition and charge transfer processes (e.g., changes in energy states of electrons bound to atoms or molecules) associated with transition metal ions such as Fe, Ti, Cr, etc., determine largely the position of diagnostic absorption features in the visible- and near-infrared wavelength region of the spectra of minerals (Burns, 1993, Adams, 1974, Adams, 1975). In addition, vibrational processes in H2O and OH− (e.g., small displacements of the atoms about their resting positions) produce fundamental overtone absorptions in the mid- to shortwave infrared part of the spectrum (Hunt, 1977). The position, shape, depth, and width of these absorption features are controlled by the particular crystal structure in which the absorbing species is contained and by the chemical structure of the material. Thus, variables characterizing absorption features can be directly related to the chemistry and structure of the sample. The absorption depth is an indicator for the amount of the material causing the absorption present in a sample. Furthermore, the absorption-band depth is related to the grain or particle-size as the amount of light scattered and absorbed by a grain is dependent on grain size. A larger grain has a greater internal path where photons may be absorbed according to Beers Law. In smaller grains there are proportionally more surface reflections compared to internal photon path lengths, if multiple scattering dominates, the reflectance decreases with increasing grain size.
Field and laboratory spectra have been used to relate absorption features to the chemical composition of samples in the areas of both soil science and mineralogy as well as in the area of vegetation science. For the analysis of hyperspectral image data there are several techniques available to derive surface composition (e.g., surface mineralogy) from a combination of absorption-band position and depth. However, no such technique provides spatial information on the variation of absorption-band depth, position and shape despite the fact that these parameters are of vital use in quantitative surface compositional mapping.
In this paper, first a review is given of the use of reflectance spectroscopy to estimate soil/rock geochemistry and foliar biochemistry using field and laboratory spectroscopy. Then, we evaluate spectral feature analysis techniques available to map absorption features in hyperspectral image data. Thereafter, a simple linear interpolation method is introduced to estimate absorption-band parameters from hyperspectral image data. Finally, in a case study the use of this technique in the analysis of airborne hyperspectral image data for surface compositional mapping is demonstrated.
Section snippets
Spectral absorption feature analysis on hyperspectral reflectance spectral data
Reflectance spectra of minerals are dominated in the visible to near-infrared wavelength range by the presence or absence of transition metal ions (e.g., Fe, Cr, Co, Ni; Hunt, 1977, Burns, 1993). The presence or absence of water and hydroxyl, carbonate and sulfate determine the absorption features in the SWIR region. The hydroxyl is generally bound to Mg or Al. The water molecule (H2O) gives rise to overtones as seen in reflectance spectra of H2O-bearing minerals. The first overtones of the OH
Analysis of spectral absorption features in hyperspectral image data
There are various techniques to process hyperspectral imagery in order to obtain surface compositional information on a pixel-by-pixel basis for the entire image (see Van der Meer et al., 2001 for a review). Techniques that specifically use absorption-band position and depth include (1) the relative absorption-band-depth (RBD) approach of Crowley et al. (1989), (2) the spectral feature fitting (SFF) technique of Clark et al. (1990a) and (3) the Tricorder (Crowley and Swayze, 1995) and
Absorption-band position, depth and asymmetry mapping from hyperspectral image data
As was discussed in previous sections, field or laboratory reflectance spectra have been used to derive compositional information on samples. Usually this involves a multiple-linear regression of absorption-band parameters and chemical composition. The following absorption-band parameters calculated from continuum removed spectra (Fig. 1) are often used: (1) the absorption-band position, (2) the absorption-band depth and (3) the absorption-band asymmetry. The relative depth, D, of the
Case study
To demonstrate the use of absorption feature mapping as described we will apply the technique to hyperspectral image data on an area of hydrothermal alteration. Here data from NASA’s Airborne Visible/Infrared Imaging Spectrometer (AVIRIS) acquired in 1995 from the Cuprite mining area situated some 30 km. south of the town of Goldfield in western Nevada are used (Fig. 3). The AVIRIS data were converted to reflectance using the empirical line calibration method described by Roberts et al. (1985).
Conclusions
Reflectance spectroscopy has been used to derive estimates of soil and rock geochemistry and foliar biochemistry using, to date, mostly field and laboratory spectroscopic techniques. These techniques most often make use of absorption feature characteristics (e.g., absorption-band wavelength position, depth and asymmetry), which are combined with geochemical analysis in a multi-linear regression to find empirical relationships with chemistry of a sample. Absorption features have also been used
Acknowledgements
I would like to thank Prof. Peter Atkinson (University of Southampton) and an anonymous reviewer for the valuable comments and suggestions that helped in improving the manuscript.
References (35)
- et al.
The reflectance spectra of organic matter in the visible near-infrared and short wave infrared region (400–2500 nm) during a controlled decomposition process
Rem. Sens. Environ.
(1997) The relationships between leaf water status, gas exchange, and spectral reflectance in cotton leaves
Rem. Sens. Environ.
(1989)- et al.
Airborne imaging spectrometer data of the Ruby mountains, Montana: mineral discrimination using relative absorption-band-depth images
Rem. Sens. Environ.
(1989) Remote sensing of foliar chemistry
Rem. Sens. Environ.
(1989)- et al.
Reflectance spectroscopy of fresh whole leaves for the estimation of chemical concentration
Rem. Sens. Environ.
(1992) - et al.
Estimating the foliar biochemical concentration of leaves with reflectance spectrometry, testing the Kokaly and Clark methodologies
Rem. Sens. Environ.
(2001) - et al.
Leaf optical properties with explicit description of its biochemical composition: direct and inverse problems
Rem. Sens. Environ.
(1996) - et al.
Spectroscopic determination of leaf biochemistry using band-depth analysis of absorption features and stepwise multiple linear regression
Rem. Sens. Environ.
(1999) - et al.
Non-destructive determination of chlorophyll content of leaves of a green and an aurea mutant of tobacco by reflectance measurements
J. Plant Physiol.
(1996) - et al.
Hyper-multispectral image analysis based on waveform characteristics of spectral curve
Adv. Space Res.
(1992)
Predicting nitrogen and chlorophyll content and concentrations from reflectance spectra (400–2500 nm) at leaf and canopy scales
Rem. Sens. Environ.
Mapping of hydrothermal alteration in the Cuprite mining district, Nevada using aircraft scanner images for the spectral region 0.46–2.36 μm
Geology
Visible and near-infrared diffuse reflectance: spectra of pyroxenes as applied to remote sensing of solid objects in the solar system
J. Geophys. Res.
Geology and mineral deposits of Esmeralda County, Nevada
Nevada Bureau of Mines Geol. Bull.
Cited by (336)
Lithological mapping of Nidar ophiolite complex, Ladakh using high-resolution data
2024, Advances in Space ResearchNew data-driven estimation of metal element in rocks using a hyperspectral data and geochemical data
2024, Ore Geology ReviewsEstimation of the distribution patterns of heavy metal in soil from airborne hyperspectral imagery based on spectral absorption characteristics
2023, Journal of Environmental ManagementEvaluating changes with vegetation cover in PRISMA's spectral, spatial, and temporal attributes and their performance for classifying savannahs in Brazil
2023, Remote Sensing Applications: Society and EnvironmentMapping depth-integrated microphytobenthic biomass on an estuarine tidal flat using Sentinel satellite data
2023, International Journal of Applied Earth Observation and Geoinformation