Research Of Recognition Method On Off-line Handwritten Chinese Characters Based On Procedure Neural Networks

Off-line Handwritten Chinese Character Recognition is one of the most important areas in the pattern recognition, and is a difficult project in computer character recognition.Due to many complex structures of Chinese characters and diverse distortion of handwritten Chinese character,the accuracy is low.It is urgent to research new method on feature extraction of Chinese CharacterThe thesis discuss the feature of Procedure Neural Networks on feature extraction of Chinese Character,The following are the major contents in the thesis:1. Analyse the structure of Chinese character,discuss the advantage and disadvantage of the structure feature which include stroke ,multiple inflection-point stroke,radical segment and character,decide to use the feature points as the base segment and give the definiens.2. discuss the advantage of Procedure Neural Networks on Recognition on Off-line Handwritten Chinese Characters.Present two kinds of models of Procedure Neural Networks to extract the feature points of Chinese character and tell the kind of the stroke.3. Training and testing of procedure neural networks.Some Off-line handwritten Chinese characters in SCUT-IRAC are selected to be training sample of Procedure Neural Networks and the appropriated network parameters are saved.A full process of Recognition on Off-line Handwritten Chinese is provided.The result of the experiment proved that relative to traditional methods,it is more effective and fast for Procedure Neural Networks used on Recognition on Off-line Handwritten Chinese Characters.