Auto-generation of the handwritten character and brush character with individuality and natural variation

2017 Fiscal Year Final Research Report

• Project/Area Number: 15K00242
• Research Category: Grant-in-Aid for Scientific Research (C)
• Allocation Type: Multi-year Fund
• Section: 一般
• Research Field: Perceptual information processing
• Research Institution: The University of Aizu
• Principal Investigator: SHIN JUNG PIL 会津大学, コンピュータ理工学部, 上級准教授 (40315677)
• Project Period (FY): 2015-04-01 – 2018-03-31
• Keywords: 筆跡 / 文字合成 / 個人性 / 変異性

Outline of Final Research Achievements

This research is concerned with the character generation method modeled personal handwriting. This research has succeeded to generate a handwritten character affected by (1) individuality and (2) natural variation of handwriting. It was based on this and 3 targets were established about this problem. First the character method to synthesize from a sample which has not written actually (called (3) scalability) has been developed aiming at realization of a more convenient system. Next the generated character was utilized and the performance of the on-line writer verification was improved. The artificial brush type character with 3 natures mentioned before was synthesized and model development for synthesis of the writing brush character was achieved.

Free Research Field

パターン認識