A fully integrated hand-printed character recognition system using artificial neural networks

The paper presents an integrated strategy for hand-printed optical character recognition. A novel image processing algorithm is proposed that enhances the low frequency features of the input data. Logical neural networks are employed to classify the data. It is recognised that the classification performance will not be error-free due to ambiguities in the data, which cannot be resolved by human interpretation. A contextual post-processor is therefore employed to provide error-correction on the recognition strings. The contextual processor uses dictionary search techniques supported by Viterbi estimators if the input string is not part of the dictionary. The system therefore is not constrained to limited vocabularies