Presentation 1999/10/22
Generation of variety of characters by changing the via-point of the handwriting based on genetic algorithm
Kei Ohkawa, Yasuhiro Wada,
PDF Download Page PDF download Page Link
Abstract(in Japanese) (See Japanese page)
Abstract(in English) In pattern recognition, a variety of characters are needed for learning like in neural network learning. As one means of collecting a variety of characters, it may be used the artificial deformed characters. We show that a computational handwriting model can be applied to the deformation of characters. The handwriting model is based on optimal principles and the representations of the model are sets of via-points extracted from actual handwritten characters. Therefore, if the information of a via-point is changed, a deformed character can be generated using the handwriting model. In this report, we propose a method of changing via-point information by a genetic algorithm. We show that the proposed method can produce various character shapes, and the characters deformed by the model can be used as a set of learning patterns.
Keyword(in Japanese) (See Japanese page)
Keyword(in English) handwriting model / via-point / deformation character / genetic algorithm
Paper # NC99-50
Date of Issue

Conference Information
Committee NC
Conference Date 1999/10/22(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 Neurocomputing (NC)
Language JPN
Title (in Japanese) (See Japanese page)
Sub Title (in Japanese) (See Japanese page)
Title (in English) Generation of variety of characters by changing the via-point of the handwriting based on genetic algorithm
Sub Title (in English)
Keyword(1) handwriting model
Keyword(2) via-point
Keyword(3) deformation character
Keyword(4) genetic algorithm
1st Author's Name Kei Ohkawa
1st Author's Affiliation Nagaoka University of Technology()
2nd Author's Name Yasuhiro Wada
2nd Author's Affiliation Nagaoka University of Technology
Date 1999/10/22
Paper # NC99-50
Volume (vol) vol.99
Number (no) 383
Page pp.pp.-
#Pages 8
Date of Issue