| Handwritten Chinese character recognition has a good prospect and high value of theory research. From the point of application, many document need put into the computer especially in the management of archives, library digitalization, envelope automatic manage. Another example is some ID checking, data list automatic reading, machine translation and so on. Many fields need a good recognition system. The whole field of character recognition has a good perspective. On the other hand as an important branch of pattern recognition and the most difficult field of character recognition, the theory and technology of traditional pattern recognition can' t take good performance in Chinese character recognition. The research in this field can help the development of the theory research in pattern recognition. Many subjects are involved such as pattern recognition image process digital process nature langue understanding artificial intelligence, fuzzy mathematics, theory of information Chinese information process, it combine many technologies . and can also help the fusion of different subjects. So it is worth to study on this field. Since Casey and Nagy, two employee of IBM company, wrote the first paper about Chinese character recognition forty years have been passed. Because of many hard work of many researchers the online recognition and print font recognition have succeed in our daily life. Only offline handwritten Chinese character still can't get good result. It is called the most difficult field in OCR, especially based on structure is a challenge for our researchers. Out of all the difficulties false connection and distortion are the most difficult. In this paper my study is based on structure and proposed a way of solute false connection. This paper propose a new stroke segment extraction algorithm, static separate radical algorithm, dynamic construct radical algorithm, refine the directional element feature, composed the statistical feature and structural feature together form... |
PDF Full Text Request