Hard and Fuzzy Segmentation Approaches

All of the methods and strategies presented in this section have a common characteristic, namely that the grouping schemes are obtained using only spectral information-that is, spectral similarity among pixels. This is a consequence of the original goal of the cluster analysis methods, oriented to find groups in spectra 

Recently, a number of modifications of the classical methods have appeared that incorporate the spatial distance among pixels as an addihonal criterion in the clustering schemes. Thus, similarity measures based on spectral distances, such as p, can be weighted incorporating pixel neighboring informahon for example, the Euclidean distance can be redefined as  

As in the earlier section, there are many works in which hard and fuzzy segmentation schemes are provided for a common problem since both furnish complementary information [92—95]. 

---