FUZZY CLASSIFICATION OF SPATIALLY DEGRADED THEMATIC MAPPER DATA FOR THE ESTIMATION OF SUB-PIXEL COMPONENTS

Fuzzy classification procedures are hypothesized to provide more class membership information than hard methods as well as specific insights about sub-pixel components. This hypothesis is investigated in the pr esent work using ancillary and spatially degraded Thematic Mapper (TM) data of a wide valley in central Italy. Hard and fuzzy modified maxim um likelihood classifications (MLC) were applied to the data set. Fuzz y membership grades were found to be related to class cover proportion s for each degraded pixel. Also, fuzzy probabilities were able to prov ide more precise estimates of class distribution, as measured by Kappa coefficients of agreement, with respect to hard attributions. Further analysis showed that this improvement was chiefly due to a better cha racterization of mixed, uncertainly attributable pixels. These results are of great interest for definition of the cases for which the fuzzy approach is appropriate.