14.

[Cover] [Contents] [Index]

Page 11

width (i.e. the region of the electromagnetic spectral from which sensor receives energy), this will cause a reduction in the spectral resolution (i.e. sensor’s ability to quantify spectral differences). Overall, it can be concluded that enhancement of both spatial and radiometric resolution cannot be achieved together, and some kind of compromise is needed.

1.2 Optical remote sensing systems

Over the past 10–15 years, the most widely used optical remote sensing systems have been the Landsat TM and MSS, the SPOT HRV and the NOAA AVHRR instruments. New remote sensing satellites carrying optical sensors have been launched recently by a number of national space agencies and private companies. Examples are the Chinese-Brazilian remote sensing system, the US Landsat-7 and Terra spacecraft, the Space Imaging Corporation’s IKONOS satellite, the Indian IRS series, and several Japanese experimental projects. Future missions will include the expansion of the U.S. EOS programme and the launch of the European ENVISAT. The EOS satellites and ENVISAT will carry a range of sensors (optical and microwave, imaging and profiling). The first private sector remote sensing satellite, IKONOS II, was launched in 1998 and will be followed by several competing systems in the next year or so. Considerable interest is also being shown in the application of hyperspectral imagery. Whereas multispectral sensors such as the Landsat ETM+ collect upwelling radiation in a small number of broad wavebands (seven in the case of the ETM+ instrument), a hyperspectral sensor collects data in a large number of very narrow wavebands. An example is the DAIS instrument, which collects data in 79 bands in the wavelength range 0.4–12.6 μm. The width of each spectral band varies from 15 to 20 nm in the visible to 2 μm in the middle infrared. The CASI instrument allows data to be collected for any 545 nm segment of the 0.4–1.0 μm region in 288 bands, with the bands spaced at intervals of approximately 1.9 nm. NASA’s AVIRIS acquires data in 224 bands in the range 0.38–1.5 μm with a bandwidth of 10 nm. Up until the present time, hyperspectral data have been collected by aircraft-mounted sensors. NASA’s Earth Observer I, launched in 2000, is the first orbiting spacecraft to carry a hyperspectral imager, Hyperion. The Hyperion instrument collects data over a narrow swath in 220 bands of 10 μm width. The large number of spectral bands produced by hyperspectral sensors (i.e. the high dimensionality of the data) poses significant problems in the pattern recognition process. These problems are discussed in Section 2.6.

An exhaustive list of sensors and satellites is not supplied here; such lists tend to become outdated very quickly. The advent of the World Wide Web (WWW) means that researchers, teachers and students can readily obtain up-to-date information via the Internet. Professional and learned societies, such as the Remote Sensing and Photogrammetry Society, maintain WWW

[Cover] [Contents] [Index]


Classification Methods for Remotely Sensed Data
Classification Methods for Remotely Sensed Data, Second Edition
ISBN: 1420090720
EAN: 2147483647
Year: 2001
Pages: 354

flylib.com © 2008-2017.
If you may any questions please contact us: flylib@qtcs.net