An improved classification for cursive script recognition
Laurent Duneau, Bernadette Dorizzi · 2002
The authors present a system able to recognize English words dynamically written on a digitizing tablet. It relies on a statistical approach, in which letters are first identified with a certain likelihood. A score is then given to each word in the dictionary for the final recognition. In order to optimize the allograph identification phase, a hierarchical classification algorithm allows one to group the allographs corresponding to a given letter and a given size. Performances on a database of 5000 words and a dictionary of 1000 words are very good and the extension to multi-scriptors is possible thanks to a partial relearning.>