Research of off-line handwritten Chinese character recognition system based on feedback structure with statistical general literacy errors

Author

Wang, Jianping ; Qiu, Jin ; Zhu, Chenghui ; Jinling Wang

Author_Institution

Sch. of Electr. Eng. & Autom., Hefei Univ. of Technol., Hefei, China

fYear

2011

fDate

16-18 Sept. 2011

Firstpage

6069

Lastpage

6072

Abstract

An off-line handwritten Chinese character recognition system based on feedback structure with statistical error of generalized literacy is constructed in this paper by imitating the feedback behavior of repeatedly considering in the human brain. Four kinds of general literacy errors are defined, which are rough error of aberration comparison, detail error of character characteristic, characteristic of detail error and matching trend error. According to the statistical analysis, rough error of aberration comparison and detail error of character characteristic is given by random statistical properties. The feedback mechanisms and decision-making strategies are adjusted by characteristic of detail error and matching trend error. The results of simulation show that this system is efficient.

Keywords

character recognition; natural languages; random processes; statistical analysis; aberration comparison rough error; character characteristic; decision-making strategy; feedback structure; matching trend error; off-line handwritten Chinese character recognition; random statistical property; statistical general literacy error; Character recognition; Decision making; Educational institutions; Feature extraction; Handwriting recognition; Humans; Indexes; closed-loop; cycle; feedback; general character recognition errors; statistical errors;

fLanguage

English

Publisher

ieee

Conference_Titel

Electrical and Control Engineering (ICECE), 2011 International Conference on

Conference_Location

Yichang

Print_ISBN

978-1-4244-8162-0

Type

conf

DOI

10.1109/ICECENG.2011.6057894

Filename

6057894