ABSTRACT
This chapter examines the major diagonal of the error matrix, measuring map class variability, and using a fuzzy error matrix approach. S. Gopal and C. Woodcock proposed the use of fuzzy sets to “allow for explicit recognition of the possibility that ambiguity might exist regarding the appropriate map label for some locations on the map. As Woodcock and Gopal state, “The assumption underlying fuzzy set theory is that the transition from membership to non-membership is seldom a step function.” The use of the fuzzy error matrix is a very powerful tool in the accuracy assessment process, because the fuzzy error matrix allows the analyst to compensate for situations where classification scheme breaks represent artificial distinctions along a continuum of land cover and/or where observer variability is often difficult to control. Using fuzzy rules in an error matrix approach combines all the established descriptive and analytical power of the error matrix while incorporating variation into the assessment.
