Glossary for Geospatial Science

Suits-Wagner classification:� A simple form of supervised or semiautomatic classification that operates similarly to a boxcar or parallelepiped classification.� This method defines the sides of the box for each class as plus and minus one standard deviation from the mean of the values of the prototype, or sample, values selected to represent that class.� This method has the advantage that like a simple boxcar classifier, it is very fast to apply the resulting decision rule to the unknown pixel values.� More information on this process can be found in: H.L. Wagner and G.H. Suits (1980) Proceedings of the 14th International Symposium on Remote Sensing of the Environment, vol. III pp. 1525�1529.� Environmental Research Institute of Michigan, Ann Arbor Michigan.� A Low Cost Classification Algorithm for Developing Countries.