Handwritten Character Recognition System

Abstract

A handwritten character recognition system is disclosed. The system has an input device for receiving handwritten strokes, and a processor for classifying the handwritten strokes as an output character. The processor does so by calculating a degree of membership of the handwritten strokes to each of a plurality of character models. The character model that produces the highest degree of membership is selected and the handwritten strokes ate classified as the output character associated with the character model that produces the highest degree of membership.

1 . A handwritten character recognition system comprising:
 an input device for receiving one or more handwritten strokes; and   a processor for:
 calculating a degree of membership of the one or more handwritten strokes to each of a plurality of character models; 
 selecting the character model that produces the highest degree of membership; and 
 classify the one or more handwritten strokes as an output character associated with the character model that produces the highest degree of membership, 
   wherein the degree of membership of the one or more handwritten strokes to each respective character model is calculated by:   segmenting the one or more handwritten strokes into sub-strokes according to a model specific segmentation scheme defined for the respective character model;   calculating a degree of conformity of each sub-stroke to a set of generic shapes defined for the respective character model;   calculating the degree of membership of the one or more handwritten strokes to the respective character model based on a set of rules specific for the respective character model and defining the positioning of the set of generic shapes.   
     
     
         2 . The handwritten character recognition system of  claim 1  wherein the input device is a graphics tablet. 
     
     
         3 . The handwritten character recognition system of  claim 1  wherein, upon failure of the segmentation of the one or more handwritten strokes into sub-strokes according to the model specific segmentation scheme defined for the respective character model, the degree of membership of the one or more handwritten strokes to the respective character model is given a low value. 
     
     
         4 . The handwritten character recognition system of  claim 1  wherein each model specific segmentation scheme defines a minimum sub-stroke length. 
     
     
         5 . The handwritten character recognition system of  claim 4  wherein the minimum sub-stroke length is based upon the length of the segment being segmented. 
     
     
         6 . The handwritten character recognition system of  claim 1 , wherein at least one character model has a plurality of associated model specific segmentation schemes. 
     
     
         7 . The handwritten character recognition system of  claim 6 , wherein the processor, for each character model having a plurality of associated model specific segmentation schemes, determines whether the one or more strokes is capable of being segmented with one of the model specific segmentation schemes, and if the one or more strokes are incapable of be segmented, the processor changes the model specific segmentation schemes. 
     
     
         6 . The handwritten character recognition system of  claim 1 , wherein fuzzy logic rules are used in calculating the degree of conformity and the degree of membership.