Presentation 2008-02-22
Handwritten Chinese Character Recognition by Background Propagation Method
Takafumi FUKUI, Michio UMEDA,
PDF Download Page PDF download Page Link
Abstract(in Japanese) (See Japanese page)
Abstract(in English) In the field of character recognition, many feature extraction methods are proposed. Those are the methods extract features from character shapes themselves like a weighted direction index histogram method. But features must exist also in the background of the character pattern. Here we propose a background propagation method as a new character recognition method which propagates the directionality of the stroke to the background. In this method, the features which are robust and stable to the position variation are obtained by using the gradation of the contour directionality and two dimensional Gaussian filter. From the recognition experiment to all characters in ETL9B, the method is effective in handwritten character recognition because the recognition rate is higher than that of the other method.
Keyword(in Japanese) (See Japanese page)
Keyword(in English) Background Propagation Method / Weighted Direction Index Histogram Method / Feature Extraction / Character Recognition / Autoassociative Neural Network
Paper # PRMU2007-233
Date of Issue

Conference Information
Committee PRMU
Conference Date 2008/2/14(1days)
Place (in Japanese) (See Japanese page)
Place (in English)
Topics (in Japanese) (See Japanese page)
Topics (in English)
Chair
Vice Chair
Secretary
Assistant

Paper Information
Registration To Pattern Recognition and Media Understanding (PRMU)
Language JPN
Title (in Japanese) (See Japanese page)
Sub Title (in Japanese) (See Japanese page)
Title (in English) Handwritten Chinese Character Recognition by Background Propagation Method
Sub Title (in English)
Keyword(1) Background Propagation Method
Keyword(2) Weighted Direction Index Histogram Method
Keyword(3) Feature Extraction
Keyword(4) Character Recognition
Keyword(5) Autoassociative Neural Network
1st Author's Name Takafumi FUKUI
1st Author's Affiliation Graduate School of Engineering, Osaka, Electro-Communication University()
2nd Author's Name Michio UMEDA
2nd Author's Affiliation Graduate School of Engineering, Osaka, Electro-Communication University
Date 2008-02-22
Paper # PRMU2007-233
Volume (vol) vol.107
Number (no) 491
Page pp.pp.-
#Pages 6
Date of Issue