Classification techniques for remotely sensed data

Hyperspectral imaging sensors measure the radiance of the materials within each pixel \narea at a very large number of contiguous spectral wavelength bands. So, they can generate \nhundreds of images of a scene on the real surface. The radiance is converted into \nhyperspectral data cube digital form. The spectral information available in a hyperspectral \nimage (cube) may serve to classify the nature of the target object because every material \nhad a unique fixed spectrum and could be used as a spectral signature of the material and \nperhaps provide additional information for further processing and exploitation. Hyperspectral \ndata contain extremely rich spectral attributes, which offer the potential to discriminate \nmore detailed classes with classification accuracy.