Remote Sensing Image Analysis: Including the Spatial Domain

 Remote Sensing Image Analysis: Including the Spatial Domain

Since the launch of the fi rst earth observation satellite ERTS-1 in 1972 much eff ort has 
been made to develop suitable and scientifi cally sound methods of information extraction  from digital images. Over the years, Remote Sensing has proven to be a valuable tool for identifying objects at the earth’s surface and for measuring and monitoring important 
biophysical characteristics and human activities on the terrain. Since the early days of earth observation, numerical methods of spectral analysis have been used to extract information from these digital images. Because computer power was limited, few spectral bands were recorded at pre-selected frequencies and visualization methods were very basic, it was only 
possible to transform the raw pixel data into meaningful classes on a pixel-by-pixel basis; one simply did not have the tools to analyse large amounts of remotely sensed digital data over a  wide range of frequencies of light.

Th e 1980s saw the development of spectral-based algorithms for image analysis and image 
classifi cation which were included in various kinds of image processing software packages. During the 1990s, spectral image analysis fl ourished anew when hyperspectral sensors with hundreds of spectral bands became available. Computers were much more powerful than before and visualization techniques had matured. Th e high spectral resolution images permitted the recording of spectral data in many spectral bands, which allowed the use of absorption feature identifi cation and the application of sub-pixel methods such as spectral mixture analysis.
Contact me: to download the book.

Post a Comment