Introduction: Hyperspectral Remote Sensing
Researchers and scientists utilize hyperspectral remote sensing, (also known as imaging spectroscopy), to identify terrestrial vegetation, minerals, and land use/land cover mapping. Remote sensing involves the examination of features observed in several regions of the electromagnetic spectrum (Remote Sensing Using the Thermal Infrared Spectrum Range). Hyperspectral remote sensing will boost remote sensing’s (The Basic Concept of Remote Sensing) analytical skills to new heights, laying the groundwork for more detailed knowledge of how to effectively build future remote sensing capabilities.
Hyperspectral Remote Sensing is based on the examination of many narrowly defined spectral channels. Sensor systems like the SPOT 1 HRV, Landsat MSS, and Landsat TM (Satellite Remote Sensing and Landsat Satellite Series) offer three, four, and seven spectral channels, respectively. The hyperspectral sensors have a very wide narrow range of spectral channels. A typical hyperspectral scanner records more than 100 bands and thus enables the construction of a continuous reflectance spectrum for each pixel. Though this data has been available since 1983, it is only now that it is becoming widely used due to a variety of complicated elements providing applications in numerous disciplines of engineering and research.
“In simple words, we may define the term hyperspectral is that it is a sensors based technology (sometimes referred to as imaging spectrometers) that acquires images in many, very narrow, contiguous spectral bands throughout the visible, near-IR, mid-IR, and thermal-IR portions of the spectrum”. These systems generally gather data in 100 or more bands, allowing the creation of a practically continuous reflectance (or emittance in the case of thermal IR radiation) spectrum for each pixel in the image. Hyperspectral sensors can generate data with sufficient spectral resolution for direct material identification, but a wider band of Landsat TM cannot resolve these diagnostic spectral variations.
Application of Hyperspectral Remote Sensing
Due to the vast number of very narrow bands recorded, hyperspectral data allow remote sensing images to data collection to replace data collection that was previously restricted to laboratory testing or costly ground site investigations.
Some application areas of hyperspectral sensing include determinations of surface mineralogy; water quality; bathymetry; soil type and erosion; vegetation type, plant stress, leaf water content, and canopy chemistry; crop type, and condition; and snow and ice properties.
Hyperspectral photography is becoming increasingly popular for monitoring environmental changes. It is often used to understand surface CO2 emissions, map hydrological patterns, and track pollution levels, among many other things.
Monitoring of Atmosphere variables such as water vapor, cloud properties, aerosols
Coastal Waters: chlorophyll, phytoplankton, dissolved organic materials, suspended sediments
Land use applications: Remotely sensed images are often analyzed using digital image processing methods such as supervised and unsupervised classification. The availability of hyperspectral data with greater spatial and spectral resolution has multiplied the possibilities for land use categorization.
Vegetation indices produced from hyperspectral sensors are more accurate and sensitive than those derived from optical images. Many applications need knowledge of the reflectance spectrum of vegetation. The leaf chemistry, which is responsible for the absorption properties of the leaf spectrum in the visible wavebands, is one of the biophysical elements that determine the spectrum of active vegetation.
In many open water aquatic habitats, hyperspectral images have been used to indirectly assess water quality by identifying the trophic status of lakes, defining algal blooms, and projecting total ammonia concentrations to monitor wetland water quality changes. The chlorophyll concentration is often assessed via remotely sensed images, which may then be used to monitor algal content and hence water quality. Because of the tiny contiguous bands, hyperspectral images may identify chlorophyll and algae more accurately.
Examples of Hyperspectral Remote Sensing Systems
The Hyperion and AC systems carried on the EO-1 spacecraft, as well as the CHRIS sensor, were among the first hyperspectral satellite sensors to be successfully launched. The EO-1 Hyperion sensor nominally offers 242 spectral bands of data ranging from 0.36 to 2.6 μm in diameter, each with a width of 0.010 to 0.011 μm. Some of the bands, particularly those towards the bottom and top of the spectrum, have a low signal-to-noise ratio.
As a result, only 198 of the 242 bands are calibrated during Level 1 processing; radiometric values in the other bands are set to 0 for most data products. This experimental sensor has a spatial resolution of 30 m and a swath width of 7.5 km. The USGS distributes data from the Hyperion system.
The EO-1 AC (also referred to as the LEISA AC or LAC) is a hyperspectral imager of coarse spatial resolution covering the 0.85- to 1.5- μm wavelength range. It was created to correct for atmospheric fluctuation caused mostly by water vapor and aerosols in imagery from other sensors. At nadir, the AC has a spatial resolution of 250 m. The following tables reflect some examples of hyperspectral remote sensing sensor systems:-
| Sensor | No of bands | Wavelength range (μm) | Bandwidth (nm) |
| CASI 1500 | 288 | 0.365–1.05 | Down to 1.9 |
| SASI 600 | 100 | 0.95–2.45 | 15 600 |
| MASI 600 | 64 | 3.0–5.0 | 32 600 |
| TASI 600 | 32 | 8.0–11.5 | 250 600 |
| AISA Eagle | Up to 488 | 0.4–0.97 | 3.3 |
| AISA Eaglet | Up to 410 | 0.4–0.97 | 3.3 |
| AISA Hawk | 254 | 0.97–2.5 | 12 |
| AISA Fenix | Up to 619 | 0.38–2.5 | 3.5–10 |
| AISA Owl | 100 | 7.6–12.5 | 100 |
| Pika II | 240 | 0.40–0.90 | 2.1 |
| Pika NIR | 145 | 0.90–1.70 | 5.5 |
| NovaSol visNIR | 120 to 180 | 0.38–1.00 | 3.3 |
| NovaSol Alpha-vis | 40 to 60 | 0.35–1.00 | 10 |
| NovaSol SWIR 640C | 170 | 0.85–1.70 | 5 |
| NovaSol Alpha-SWIR | 160 | 0.90–1.70 | 5 |
| NovaSol Extra-SWIR | 256 | 0.86–2.40 | 6 |
| AVIRIS | 224 | 0.4–2.5 10 677 | 10 |
| Probe-1 | 128 | 0.4–2.5 | 11–18 |
The very first commercially made available hyperspectral scanner was the Compact Airborne Spectrographic Imager (CASI) which collected data using 288 bands between 0.4 and 0.9 μm with an instantaneous field of view of 1.2 mrad(Milliradian). This technique was used in association with the global positioning system GPS (Global Position System: Different Segments of GPS, its working Principle, Popular Substitute of GPS) to correct for fluctuations in aircraft altitude. The Advance Airborne Hyperspectral Imaging Spectrometer (AAHIS) is another commercially available hyperspectral scanner that gathers data in about 288 channels between 0.40 and 0.90 μm.
Recommended Books for the Detail Study of Hyperspectral Remote Sensing
