Presentation 2007-11-29
Generation of Handprinted Characters Based on Direction-Array Analysis
Kei KUREISHI, Fujio TORIUMI, Kenichiro ISHII,
PDF Download Page PDF download Page Link
Abstract(in Japanese) (See Japanese page)
Abstract(in English) One of the serious problems in developing a character recognition system is designing a recognition dictionary which requires a tremendous amount of character data for training. A traning set, however, contains only limited variations of characters. The purpose of this study is to generate various kinds of distorted characters artificially. We propose a Direction-Array which represents a character as a sequence of minimum line segments. We also propose distorted character generation using the weighted sum of a pair of Direction-Arrays. In the experiment, the recognition result indicated that the dictionary designed by artificially distorted characters has almost the same performance as the dictionary designed by actual characters. It is also shown that the proposed method can evaluate the stability of the recognition systems against distortion.
Keyword(in Japanese) (See Japanese page)
Keyword(in English) Direction-Array / Handprinted Character / Character Generation / Online Character
Paper # PRMU2007-133,MVE2007-66
Date of Issue

Conference Information
Committee PRMU
Conference Date 2007/11/22(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) Generation of Handprinted Characters Based on Direction-Array Analysis
Sub Title (in English)
Keyword(1) Direction-Array
Keyword(2) Handprinted Character
Keyword(3) Character Generation
Keyword(4) Online Character
1st Author's Name Kei KUREISHI
1st Author's Affiliation Graduate School of Information Science, Nagoya University()
2nd Author's Name Fujio TORIUMI
2nd Author's Affiliation Graduate School of Information Science, Nagoya University
3rd Author's Name Kenichiro ISHII
3rd Author's Affiliation Graduate School of Information Science, Nagoya University
Date 2007-11-29
Paper # PRMU2007-133,MVE2007-66
Volume (vol) vol.107
Number (no) 359
Page pp.pp.-
#Pages 6
Date of Issue