Spectral Angle Mapper

Home Image Analysis Classification Spectral Angle Mapper

Index	Statistical Background
Spectral Angle Mapper The Spectral Angle Mapper (SAM) classification approach is based on the idea looking at spectra as p-dimensional vectors. The more similar two spectra are the smaller is the angle between the two spectral vectors. The SAM approach has the advantage that the results are insensitive to illumination because only the direction of the spectral vector is used and not its length.