Presentation | 1998/2/19 An Online Handwritten Character Segmentation Method of which Parameters can be Decided by Learning Shuji SENDA, Masahiko HAMANAKA, Keiji YAMADA, |
---|---|
PDF Download Page | PDF download Page Link |
Abstract(in Japanese) | (See Japanese page) |
Abstract(in English) | In this report, a character segmentation method is proposed which obtains shape feature vectors from segmentation candidates, and evaluates the candidates by linear transformation of their feature vectors, and searches a segmentation path in which the sum of evaluated values is the best. The advantage of the proposed method is that parameters for the linear transformation can be optimized by a steepest gradient method so that segmentation rate of training data should be the best. Experiments show this method effective. |
Keyword(in Japanese) | (See Japanese page) |
Keyword(in English) | online handwritten character segmentation / segmentation parameters / optimization |
Paper # | |
Date of Issue |
Conference Information | |
Committee | PRMU |
---|---|
Conference Date | 1998/2/19(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) | An Online Handwritten Character Segmentation Method of which Parameters can be Decided by Learning |
Sub Title (in English) | |
Keyword(1) | online handwritten character segmentation |
Keyword(2) | segmentation parameters |
Keyword(3) | optimization |
1st Author's Name | Shuji SENDA |
1st Author's Affiliation | NEC C&C Media Research Laboratories() |
2nd Author's Name | Masahiko HAMANAKA |
2nd Author's Affiliation | NEC C&C Media Research Laboratories |
3rd Author's Name | Keiji YAMADA |
3rd Author's Affiliation | NEC C&C Media Research Laboratories |
Date | 1998/2/19 |
Paper # | |
Volume (vol) | vol.97 |
Number (no) | 558 |
Page | pp.pp.- |
#Pages | 8 |
Date of Issue |