Proceedings of the 2017 2nd Joint International Information Technology, Mechanical and Electronic Engineering Conference (JIMEC 2017)

A Novel Blocked Winner Sequence Feature Extraction Method Based SOM For Large Set Chinese Character Recognition

Authors
Lei Wang, Wenjuan Zhang, Lianming Wang
Corresponding Author
Lei Wang
Available Online October 2017.
DOI
10.2991/jimec-17.2017.4How to use a DOI?
Keywords
feature extraction; blocked winner sequence; SOM neural networks; Chinese characters recognition; large character set.
Abstract

A novel feature extraction method named blocked winner sequence feature extraction (BWS) method based on SOM is proposed for recognizing large set of Chinese characters. Firstly, a Chinese character sample is blocked to make sub block set according to certain principles; Secondly, every sub-block is passed into a SOM network orderly for learning, and the location number of the winner for each sub-block is stored; Finally, location numbers of all winner neurons are combined into a sequence as the feature of the character. Compared with traditional SOM neural networks, the method using orderly combination of multiple winner neurons instead of single winner to represent feature can reduce the network size and computation as well as improve network capacity effectively. Using block can improve the anti-interference ability and the recognition accuracy of the network. The method is used for the extract feature of 3500 Chinese characters in GB2312 Chinese character library with Euclidean distance recognition, and the recognition accuracy reaches 99.897% and 93.274% respectively in the cases of adding 10% and 20% random noise.

Copyright
© 2017, the Authors. Published by Atlantis Press.
Open Access
This is an open access article distributed under the CC BY-NC license (http://creativecommons.org/licenses/by-nc/4.0/).

Download article (PDF)

Volume Title
Proceedings of the 2017 2nd Joint International Information Technology, Mechanical and Electronic Engineering Conference (JIMEC 2017)
Series
Advances in Computer Science Research
Publication Date
October 2017
ISBN
10.2991/jimec-17.2017.4
ISSN
2352-538X
DOI
10.2991/jimec-17.2017.4How to use a DOI?
Copyright
© 2017, the Authors. Published by Atlantis Press.
Open Access
This is an open access article distributed under the CC BY-NC license (http://creativecommons.org/licenses/by-nc/4.0/).

Cite this article

TY  - CONF
AU  - Lei Wang
AU  - Wenjuan Zhang
AU  - Lianming Wang
PY  - 2017/10
DA  - 2017/10
TI  - A Novel Blocked Winner Sequence Feature Extraction Method Based SOM For Large Set Chinese Character Recognition
BT  - Proceedings of the 2017 2nd Joint International Information Technology, Mechanical and Electronic Engineering Conference (JIMEC 2017)
PB  - Atlantis Press
SP  - 16
EP  - 21
SN  - 2352-538X
UR  - https://doi.org/10.2991/jimec-17.2017.4
DO  - 10.2991/jimec-17.2017.4
ID  - Wang2017/10
ER  -