Learning templates from fuzzy examples in structural pattern recognition

A fuzzy-attribute graph (FAG) has been proposed to handle fuzziness in the pattern primitives in structural pattern recognition. FAG has the advantage that one can combine several possible definitions into a single template. However, the template requires human expert definition. In this paper, the author proposes an algorithm that can, from a number of fuzzy instances, find a template that can be matched to the patterns by the original matching metric.<
>