LARS Tech Report Number
111572
Abstract
Pattern recognition plays a central role in numerically oriented remote sensing systems. It provides an automatic procedure for deciding to which class any given ground resolution element should be assigned. The assignment is made in such a manner that on the average correct classification is achieved.
This information note describes briefly the theoretical basis for the pattern-recognition-oriented algorithms used in LARSYS, the multispectral data analysis software system developed by the Laboratory for Applications of Remote Sensing (LARS).
Date of this Version
January 1972
COinS