Fuzzy feature description of handwriting patterns

Ashutosh Malaviya and Liliane Peters GMD German National Research Center for Information Technology Schloß Birlinghoven, 53754 St. Augustin, Germany [email protected] http://set.gmd.de/EIA/fohres.html Abstract: The handwritten characters differ not only from person to person but also from the state of mood of the same person. Nevertheless humans are trained to extract the specific features characterizing a symbol. We introduced fuzziness in the definition of the proposed pattern features, which provides the enhancement to the handwritten character information to be stored. Some novel shape features in fuzzy linguistic domain are described. We conclude the paper by presenting some experimental results from the fuzzy on-line handwriting recognition system FOHRES.