INVARIANT HANDWRITTEN CHINESE CHARACTER-RECOGNITION USING FUZZY RING DATA

An invariant handwritten Chinese character recognition system is propo sed. Characters can be in arbitrary location, scale and orientation. F ive invariant features are employed in this study. The first four feat ures are used for preclassification to reduce matching time. The last feature, ring data, constructs ring-data vectors to characterize chara cter samples and constructs weighted ring-data matrices to characteriz e characters to further reduce matching time. Fuzzy membership functio ns are defined based on these two characteristics to match characters. A character set is constructed from 200 handwritten Chinese character s and comprising several different samples of each character in arbitr ary orientations. The performance of the proposed invariant features a nd fuzzy matching is verified through extensive experiments with the c haracter set: (i) the performance of the proposed fuzzy matching is su perior to that of two traditional statistical classifiers; (ii) the pe rformance of the fuzzy ring-data vector is clearly superior to that of the fuzzy ring-data matrix, but the latter needed less matching time; (iii) the preclassification reduces the fuzzy matching time and impro ves the recognition rate; and (iv) the performance of the proposed inv ariant features is clearly superior to that of moment invariants.